Abstract Sixty years of Operational Research

* Corresponding author. Tel:(98)(21)22010690./fax: (98)(21)22043001. E-mail addresses: sadinejad@ (S. Sadi-Nezhad),© 2010 Growing Science Ltd. All rights reserved. doi: 10.5267/j.ijiec.2010.03.006International Journal of Industrial Engineering Computations 2 (2011) 167–178Contents lists available at GrowingScienceInternational Journal of Industrial Engineering Computationshomepage: /ijiecPeriodic and continuous inventory models in the presence of fuzzy costsSoheil Sadi-Nezhad a*, Shima Memar Nahavandi a and Jamshid Nazemi aaDepartment of Engineering, Science & Research Branch, Islamic Azad University, Tehran IranA R T I C L E I N F O AB S T R AC TArticle history :Received 10 January 2010 Received in revised form 22 June 2010Accepted 24 June 2010Available online 25 June 2010This paper presents two models, a periodic review model and a continuous review inventory model with fuzzy setup cost, holding cost and shortage cost. We use two methods in the name of signed distance and possibilistic mean value to defuzzify. Also we consider the lead time demand and the lead-time plus one period’s demand as random variables. To validate the models and the solution procedures we apply them to a transformer manufacturing, 'Iran transfo', company. Furthermore we design a decision support system which can be used for efficient evaluation of the proposed models in fuzzy environment.© 2010 Growing Science Ltd. All rights reserved.Keywords :Fuzzy inventoryPeriodic review inventory model Continuous review inventory modelSigned distance method1. IntroductionThe increase in global competition especially among the manufacturing sectors motivates many contemporary organizations to spend more efforts for optimizing their global supply chain. Managing inventory as a primary function of supply chain, in an efficient and effective way, plays an important role on reducing the total cost of supply chain. The main purpose of the inventory management practices in all production companies is to have the required items ready in the predefined schedule with incurring minimum cost (Cakir & Canbolat, 2008). The proper control of their levels usually brings significant savings in costs. The development of inventory theory has evolved since the early 1920s, started with basic models with a few parameters to describe the deterministic components (Kao & Hsu, 2002). However, in the real world applications, estimating the precise value of parameters like demand and costs may not be possible. What a decision maker faces in this case is a fuzzy environment. In this regard, the preliminary models are changed to include more details using probabilistic models and fuzzy models, to overcome this limitation.Fuzzy set theory introduced by Zadeh (1963), provides an alternate to handle the vague situations. It has been used in modeling of inventory systems with vague and imprecise parameters since 1980s. In recent years, various types of inventory problems have been developed in fuzzy environments. Yao and Lee (1999) introduced a backorder inventory model with fuzzy order quantity as triangular and trapezoidal fuzzy numbers and shortage cost as a crisp parameter. Yao and Su (2000) developed fuzzy inventory model without backorder by adopting an interval valued fuzzy number for fuzzifying the168total demand quantity. Chang et al. (2004) proposed mixture inventory model involving variable lead time with backorders and lost sales by introducing the fuzziness of lead time demand, the average demand per year and the backorder rate of the demand during the stock-out period. Yao and Chiang (2003) considered the total cost of inventory without backorder. They fuzzified the total demand and cost of storing one unit per day into triangular fuzzy numbers and defuzzified by the centroid and the signed distance methods. Lin (2008) developed the inventory problem for a periodic review model with variable lead time and fuzzified the expected demand shortage and backorder rate using signed distance method to defuzzify. Wu and Yao (2003) fuzzified the order quantity and shortage quantity into triangular fuzzy numbers in an inventory model with backorder and they obtained the membership function of the fuzzy cost and its centroid. Handfield et al. (2009) developed a (Q , r ) model based on fuzzy-set including fuzzy parameters like demand, lead time, supplier yield, and shortage cost. They also introduced a human risk attitude factor to quantify the decision maker’s attitude toward the risk of stocking out during the replenishment period. The total cost was computed using centroid as defuzzification method. Tütüncü et al. (2009) used fuzzy set concepts to treat the uncertainty regarding costs in models of continuous review inventory control with or without backorder, with fuzzy costs and probabilistic demand then used Park's Median Rule to defuzzify. Vijayan and Kumaran (2008) introduced stochastic (Q , r ) and (R ,T ) inventory models with a mixture of backorders and lost sales. They introduced fuzziness in the cost parameters to study the impact and sensitiveness of the impreciseness of cost components in the decision variables by trapezoidal fuzzy numbers and adopted the signed distance method to estimate the fuzzy cost function of the models. Dey and Chakraborty (2009) developed a periodic review model in a mixed imprecise and/or uncertain environment in which the customer demand, lead-time demand and the lead-time plus one period’s demand are assumed to be fuzzy random variables. The fuzzy expected cost is defuzzified by its possibilistic mean value. Table1 shows a summary of literature review.Table 1The summery of literature reviewDefuzzification MethodFuzzy Parameters Year Author Centroid, Signed distancemethodtotal demand, holding cost2003 Yao and Chiang centroidorder quantity, shortage quantity 2003 Wu and Yao Signed distance methodtotal demand, lead time demand,backorder rate2004 Chang and Yao Median rulesetup cost, holding cost, shortage cost,unit cost 2007 Tutunchu and Akoz Signed distance methodsetup cost, holding cost, shortage cost,lost sale cost2008 Vijayan and Kumaran Signed distance methodexpected demand shortage, backorderrate2008 Lin Possibilistic mean valuetotal demand, lead-time demand and thelead-time plus one period’s demand 2009 Dey and Chakraborty centroiddemand, lead time, supplier yield, andshortage cost 2009 Handfield, Warsing Signed distance method,Possibilistic mean value setup cost, holding cost, shortage cost2010 Sadinejad and MemarDue to some fluctuations in inventory costs, these parameters are described as “approximately equal to”. Therefore, in this research we characterize these parameters as fuzzy variables in a periodic model and a continuous inventory model with backorder. Yao & Chiang (2003) introduced the vantage of signed distance method comparing to centroid method and we prefer to use this method.S. Sadi-Nezhad et al./ International Journal of Industrial Engineering Computations 2 (2011)169Also for the ease of computation we use Possibilistic mean value as another method for deffuzification (Day & Chakraborty, 2009).This article is organized as follows: Section 2 reviews some definitions and properties about fuzzy sets and defuzzification methods which would be used later. Section 3, deals with the methodology where the models and their optimal solutions are discussed. The subsection 3.1 explores the inventory problem for a periodic review and subsections 3.1.1 and 3.1.2 describes the defuzzified model by signed distance method and possibilistic mean value method. Section 3.2 introduces the continuous review inventory model in which section 3.2.1 and 3.2.2 are related to its defuzzification through signed distance and possibilistic mean value methods, respectively. Section 4 shows the results of the models for and Iranian case study called "Iran Transfo". In section 5, a decision support system with a user-friendly interface to obtain efficient solution of the proposed models is presented. Finally, conclusion remarks are given at the end to summarize the contribution of the paper.2. Fuzzy preliminariesFuzzy set theory provides an appropriate framework to treat imprecision of modeling uncertainties and it offers more flexibility in describing these uncertainties.Definition 1. A fuzzy set a~ on ),(+∞−∞=R is called a fuzzy point if its membership function is as follows,⎩⎨⎧≠==a x ax x a 01)(~μwhere a is called its support.Definition 2. The fuzzy set A ~= (a,b,c) is called the triangular fuzzy number where a < b < c , if the membership function of A is given by⎪⎪⎪⎩⎪⎪⎪⎨⎧≤≤−−≤≤−−=.0)(~otherwise c x b b c x c b x a a b a x x AμDefinition 3. The cut −αof A ~= (a,b,c) where 10≤≤α is: )](),([)(αααR L A A A =.)(αR A and )(αL A are the left and right end points of )(αA and are defined as follow, )(a b a A −+=−αα , )(b c c A −−=+αα(1)Property1. For the fuzzy triangular numbers A ~= (a,b,c) and ),,(~r q p B =we have,()(,,)()(,,)A B a p b q c r A B a p b q c r +=+++−=−−−%%%% (,,),0(.)(,,),00,0ka kb kc k k A kc kb ka k k >⎧⎪=<⎨⎪=⎩%170Definition 4. The defuzzification of A ~can be found by centroid or signed distance methods. Thecentroid of A ~ is 3)~(c b a A C ++=and the signed distance from A ~ to 0 is defined as follows,[]()420~,)(),()0~,~(10c b ad A AdA d RL ++==∫ααα(2)According to Kumaran (2007), it can be concluded that using )0~,~(A d to defuzzify a fuzzy number, isbetter than using )~(A C .2.1 Possibilistic mean value of a fuzzy numberFor a given fuzzy number A ~, the interval-valued possibilistic mean is defined as,())[(),()]L R M A M A M A =%%%,where )~(A M L and )~(A M U are the lower and the upper values of A ~and are, respectively, defined by Day and Chakraborty (2009).∫∫−=110)~(αααααd d A A M L and ∫∫+=110)~( αααααd d A A M RThe possibilistic mean value of A ~is then defined as,ααααd A A A M )()~(1+−+=∫(3)3.MethodologyWe make use of the following notations through this paper:T the time between replenishments; R target inventory level; Q lot size in units; r reorder point;A unit cost of placing an order for an item; h inventory holding cost per unit per year; π unit shortage cost; D annual demand3.1 periodic review inventory modelA periodic review system involves determining the amount of an item in stock at a specified, fixed time interval and placing an order that when added to the quantity on hand, will equal to predetermined maximum level. The target inventory level R , is the sum of expected demand during the lead time and replenishment period plus the safety stock. Due to fluctuations in demand orS. Sadi-Nezhad et al./ International Journal of Industrial Engineering Computations 2 (2011)171variable lead time, stock out may occur, in which case a penalty per unit shortage is incurred. The purpose of the periodic inventory model is to find the optimal time between replenishment and maximum inventory level such that the combination of the order cost and the holding cost and shortage cost is minimized. The approximate mean total cost is expressed as follows,),,(2),(T R B TDT DL R h TA T R C π+−−+=⎥⎦⎤⎢⎣⎡(4)where ),(T R B is the expected shortage at the end of the cycle and is defined as follows,,)()()(),(dx x f R x R D E T R B RT L ∫∞++−=−=(5)where x denotes the lead-time plus one period’s demand that follows a normal distribution. )(x f is normal probability distribution function with mean T L +μand standard deviation L T +σ. To minimize the C(R,T) we solve 0)0),,((=∂∂RT R C d which yields,πhTdx T x f R∫∞=),(.(6)The complementary cumulative distribution of x is as follows,⎟⎠⎞⎜⎝⎛−=−πhT F R 11. (7)The initial value of T concludes the value of R and consequently the value of C(R,T) can be calculated. This procedure is iterated until the condition 0),()1,1(>−−−i T i R C i T i R C is achieved in stage i , which means the cost needs to be decreased until it attains its minimum which yields *T and *R to be 1*−=i T T and 1*−=i R R .Note that the values of T L +μand T L +σ must be recalculated in the cumulative distribution function for each iteration.3.1.1 Fuzzy periodic review inventory model using signed distance methodSince evaluating actual costs in real world is very difficult, if not impossible, we consider all costs in the model in fuzzy sense. Consider the fuzzy costs as follows,),,(21λππλππ+−=,),,(21Δ+Δ−=h h h h ,),,(21δδ+−=A A A A . (8)By substituting these fuzzy costs into (4) we get three expected annual costs as follows,()()2)((),(1111T R D E DTDL R h T A T R C T L λπδ−−+−−Δ−+−=++,T R D E DT DL R h T A T R C T L π++−+⎥⎦⎤⎢⎣⎡−−+=)(2),(2, (9))()(2)((),(2223TR D E DT DL R h T A T R C T L λπδ+−+−−Δ+++=++.172Using the signed distance method given in (2) yields the following defuzzified value of C(R,T),().),((2(41)0),,((1212122⎥⎦⎤⎢⎣⎡−+−−Δ−Δ+−+=T R B T DT DL R T C T R C d λλδδSimilar to the crisp case, the optimal values of R and T are obtained by taking the derivative of C(R,T) with respect to R , which yields,44),(5634πππ−+⎥⎦⎤⎢⎣⎡Δ−Δ+=∫∞T h dx T x f R , or⎟⎟⎟⎟⎠⎞⎜⎜⎜⎜⎝⎛+⎥⎦⎤⎢⎣⎡Δ−Δ+−=−44156341πππT h F R .(10)The solution procedure proposed for the crisp case in the previous section is implemented here to find *T and *R .3.1.2 Fuzzy periodic review inventory model using possibilistic mean value methodThe aim of this section is to use the −αcut of the fuzzy inventory costs to derive the −αcut of the fuzzy total cost. The −αcuts of the setup cost, holding cost and shortage cost are calculated using (1). Substituting the left endpoints of the costs in (4) results the left endpoint of the total cost and the right endpoints results the right endpoint of the total cost as follow,.),(2,),(2T T R B DT DL R h T A C T T R B DT DL R h T A C ++++−−−−+⎥⎦⎤⎢⎣⎡−−+=+⎥⎦⎤⎢⎣⎡−−+=ααααααααππ (11)The possibilistic mean value of the above interval through (3) is determined as follows,()αααd C C T R C ∫+−+=1),(.The procedure to find the optimal values is the same as section 3.1 but R is determined as follows where u is defined as,πhTu −=1,(12)with 0=α, the two lower edges and with 1=α, the upper edge of the triangle are made as follows,S. Sadi-Nezhad et al./ International Journal of Industrial Engineering Computations 2 (2011)1731311πT h u −=, 2221πT h u −=, 3131πTh u −=.(13)The left and the right endpoints of this interval are as follows,).(),(233121u u u u u u u u −−=−+=+−αααα(14)Therefore, the defuzzified value using its possibilistic mean value given in (3) is calculated as follows,(),10αααd u u u ∫+−+=(15)and R is also calculated as )(1u F R −=.3.2 Continuous review inventory systemIn a continuous review inventory system, an order for a lot is placed whenever the quantity on hand is dropped to a predetermined level, known as the order point. The total cost expression of this model used in this paper is presented as follows,,)(2),(⎟⎟⎠⎞⎜⎜⎝⎛+⎥⎦⎤⎢⎣⎡−++=r B Q Dr Q h Q AD r Q C πμ(16)where )(r B is the expected demand shortage at the end of the cycle and is defined as follows,,)()()()(∫∞+−=−=RL dx x f r x r D E r B(17)where x is regarded as the value of lead-time demand with normal probability distribution function )(x f , mean L μ and standard deviation L σ. Taking the first derivative of C(Q,r) with respect to Q and r yields the optimal values as follows,,)]([2*hr B A D Q π+=(18)u 3174.11*⎟⎠⎞⎜⎝⎛−=−D hQ F r π(19)3.2.1 Fuzzy continuous review inventory model using signed distance methodBy substituting fuzzy parameters in (16) and getting three total costs like the periodic model, the defuzzified cost using signed distance method is determined as follows,⎥⎦⎤⎢⎣⎡⎟⎟⎠⎞⎜⎜⎝⎛−+⎥⎦⎤⎢⎣⎡−+Δ−Δ+−+=)()(2)()(41),()0),,((121212r B Q Dr Q QDr Q C r Q C d λλμδδ.Taking the partial derivatives with respect to Q and r yields,,4)()]()4)((4)([2121212*Δ−Δ+−++−+=h r B A D Q λλπδδ (20)and⎟⎟⎟⎟⎠⎞⎜⎜⎜⎜⎝⎛−+Δ−Δ+−=−D Q h F r )4)(()4)((1 12121*λλπ.(21)Since the optimum values of Q and r cannot be directly obtained by Eq. (20) and Eq. (21), the iterative procedure suggested by Hadley and Whitin (1963) is used to solve the equations. The initial value for Q is obtained by equating B(r) to zero and with the r resulted from substituting Q in (21) and B(r) is obtained by substituting r in (17) and the new Q is obtained by Eq. (20). Repeating this procedure until the condition 1−=i i Q Q is met, will lead us to find the optimal values of Q and r .3.2.2 Fuzzy continuous review inventory model using possibilistic mean value methodSimilar to section 3.1.2, the cut −α approach is used to treat the fuzziness and derive the total annual cost as follows,⎟⎟⎠⎞⎜⎜⎝⎛+⎥⎦⎤⎢⎣⎡−++=⎟⎟⎠⎞⎜⎜⎝⎛+⎥⎦⎤⎢⎣⎡−++=++++−−−−)(2),()(2),(r B Q D r Q h QD A r Q C r B Q D r Q h Q D A r Q C ααααααααπμπμ(22)()αααd C C r Q C ∫+−+=1),(.The minimum, the mean and the maximum values of the Q represented by 1Q ,2Q and 3Q , respectively, are calculated as follows,3111)]([2h r B A D Q π+= 2222)]([2h r B A D Q π+=1333)]([2h r B A D Q π+=S. Sadi-Nezhad et al./ International Journal of Industrial Engineering Computations 2 (2011)175The possibilistic mean using (3) is then presented as follows,().1αααd Q QQ ∫+−+=(23)Repeating the same concept applied in section 3.1.2 we have,D Q h u 1311π−=, ,1222D Q h u π−= D Q h u 3131π−=, (),1αααd u u u ∫+−+= )(1u F r −=. (24)The iterative process in section 3.2.1 is applied here to find optimal Q and r . Fig.1. gives an overview of the inputs and outputs of the models.Inputs ModeloutputsFig .1. Inputs and outputs of the models4. Application in Iran TransfoIn the previous sections, we have shown how to modify the crisp models for continuous and periodic review inventory systems in order to consider the fuzziness associated with the costs. Recently, there has been an increase interest on using fuzzy theory for real-world applications (Modarres et al. (2010). In order to validate and illustrate the results of the proposed models we have conducted an application in a private company called "Iran Transfo". We have also selected the most strategic items which constitute group products A and B in ABC inventory classification. The total demand, the lead time and the fuzzy costs, as inputs, are shown in Table 2. Deficiency of any of these items leads to an interruption on manufacturing which causes approximately 35,000,000 US dollar. The results for both models, with both defuzzification methods are determined in Table 3.176Table2Input parametersnameTotaldemandLead time(month)Fuzzy setup cost Fuzzy holding cost1 4.488 4 (95000, 100,000, 103,000) (1000, 1500, 1800)2 3240 1 (9700 10000, 10200)(115,120,122)3 960 5 (118,000, 120,000, 121,000) (580,600,620)4 10800 6 (67,000, 70,000, 72,000) (285,300,310)5 103206 (56,000, 60,000, 62,000) (200,225,250)Table3The optimal solutionsPeriodic model results Continuous model results Fuzzymethod C(R,T) R T C(Q,r) Q r1 SDM 2,556,900 3078 54 2,288,800 784 2229α-cut 2,578,500 3077 54 2,308,100 790 22282 SDM 159,720 1503 72 125,660 736 576α-cut 159,990 1504 72 125,870 735 577 3 SDM 3,049,900 9621 54 2,634,900 1934 7622α-cut 2,975,400 1023 54 2,621,700 1832 76294 SDM 1,544,900 9867 72 1,399,100 2300 7336α-cut 1,547,200 9867 72 1,370,900 2195 7341 5 SDM 262,680 1266 216 230,280 644 577α-cut 272,630 1265 216 238,940 6345765. Decision support systemIn today’s timeliness production environment, it is extremely important for the decision makers to have access to the decision support tools in order to make rapid and accurate decisions. In this section a decision support system is developed to efficiently use the models and corresponding iterative procedures and to analyze the effectiveness of the proposed approaches when tackling real world problem instances. The technological background of the system is illustrated in Fig. 2.Fig. 2. The structure of DSSS. Sadi-Nezhad et al./ International Journal of Industrial Engineering Computations 2 (2011)177The proposed DSS model consists of three major components where the first one is a database component in which all inventory records are stored and can be modified for updates. The database is implemented with Microsoft access. The second component is a model based component which includes both continuous and periodic review models with user-defined parameters. The models and the procedures are written in MATLAB software. The third item is the user interface which is designed in Visual Basic. Transforming data between MATLAB and Visual Basic is performed by using an excel file as an intermediate to pass data between these two environments. The main interface window of DSS is given in Fig. 3. As we can observe, user can see the results for both models and choose the proper one according to their condition. Also user can choose the time period and see the optimal result foe from the selected period. The point is that both models perform the calculation using both defuzzification methods and the results for the method which yields the minimum cost are shown in the form.Fig. 3. The main form of the DSS6. ConclusionIncorporating fuzzy set theory and probability theory into inventory models potentiates the model to tackle real world fuzziness and randomness. In this article, we have applied the fuzzy set theory to reformulate the two periodic and continuous inventory models where shortages are backordered with shortage cost are incurred. Also, we have used the probabilistic theory to express the demand in the replenishment time. In both models, we have used the triangular fuzzy numbers to represent the imprecise setup cost, holding cost and shortage cost and obtained the model with fuzzy total cost. The178periodic model gives the optimal values of review periods and maximum inventory levels and the continuous review model gives the optimal order quantity and order point. For these fuzzy models, we have employed two methods of defuzzification which are the signed distance and the possibilistic mean value. The implementation of the proposed method has also been illustrated using some numerical example. In addition, we have also proposed a decision support system with a user friendly interface, for efficient and effective use of the proposed models. Hence, the suggested inventory models are observed to be executable and useful for the decision maker in the real world. ReferencesCakir, O. & Canbolat, M. S. (2008). A web-based decision support system for multi-criteria inventory classification using fuzzy AHP methodology, Canada Expert Systems with Applications, 35, 1367–1378.Chang, H-C., Yao, J-S. & Ouyang, L-Y. (2004). Fuzzy mixture inventory model with variable lead-time based on probabilistic fuzzy set and triangular fuzzy number, Mathematical and Computer Modelling, 39, 287-304.Day, O. & Chakraborty, D. (2009). Fuzzy periodic review system with fuzzy random variable demand. European Journal of Operational Research, 198(1), 113-120.Kao, C-K & Hsu, W-K, (2002). Lot size-reorder point inventory model with fuzzy demands, Computers & Mathematics with Applications, 43, 1291-1302.Hadley, G., Whitin, T.M., (1963). Analysis of Inventory Systems. Prentice-Hall, New Jersey. Handfield, R., Warsing, D. & Wu, X. (2009). (Q,r) Inventory policies in a fuzzy uncertain supply chain environment, European Journal of Operational Research, 197(2), 609-619.Lin, Y-J. (2008). A periodic review inventory model involving fuzzy expected demand short and fuzzy backorder rate, Computers & Industrial Engineering, 54(3), Pages 666-676.Modarres, M, Sadi-Nezhad, S & Arabi, F. (2010), Fuzzy analytical hierarchy process using preference ratio: A case study for selecting management short course in a business school, International Journal of Industrial Engineering Computations, 1(2), 173-184.Tütüncü, G. Y., Aköz, O., Apaydın, A. & Petrovic, D. (2008). Continuous review inventory control in the presence of fuzzy costs, International Journal of Production Economics, 113(2),775-784. Vijayan, T. & Kumaran, M. (2008). Inventory models with a mixture of backorders and lost sales under fuzzy cost, European Journal of Operational Research, 189(1), 105-119.Wu, K. & Yao, J-S., (2003). Fuzzy inventory with backorder for fuzzy order quantity and fuzzy shortage quantity, European Journal of Operational Research, 150(2), 320-352.Yao, J-S. & Lee, H-M. (1999). Fuzzy inventory with or without backorder for fuzzy order quantity with trapezoid fuzzy number, Fuzzy Sets and Systems, 105(3), 311-337.Yao, J-S. & Su, J-S. (2000). Fuzzy inventory with backorder for fuzzy total demand based on interval-valued fuzzy set, European Journal of Operational Research, 124(2), 390-408.Yao, J-S. & Chiang, J. (2003). Inventory without backorder with fuzzy total cost and fuzzy storing cost defuzzified by centroid and signed distance, European Journal of Operational Research, 148(2, 401-409.Zadeh, L., (1965). Fuzzy sets, Information and Control, 8, 338-353.。

Linear Combinations of Optic Flow Vectors for Estimating Self-Motion–a Real-World Test of aNeural ModelMatthias O.FranzMPI f¨u r biologische KybernetikSpemannstr.38D-72076T¨u bingen,Germany mof@tuebingen.mpg.deJavaan S.ChahlCenter of Visual Sciences,RSBSAustralian National UniversityCanberra,ACT,Australiajavaan@.au AbstractThe tangential neurons in thefly brain are sensitive to the typical opticflow patterns generated during self-motion.In this study,we examinewhether a simplified linear model of these neurons can be used to esti-mate self-motion from the opticflow.We present a theory for the con-struction of an estimator consisting of a linear combination of opticflowvectors that incorporates prior knowledge both about the distance distri-bution of the environment,and about the noise and self-motion statisticsof the sensor.The estimator is tested on a gantry carrying an omnidirec-tional vision sensor.The experiments show that the proposed approachleads to accurate and robust estimates of rotation rates,whereas transla-tion estimates turn out to be less reliable.1IntroductionThe tangential neurons in thefly brain are known to respond in a directionally selective manner to wide-field motion stimuli.A detailed mapping of their local motion sensitivities and preferred motion directions shows a striking similarity to certain self-motion-induced flowfields(an example is shown in Fig.1).This suggests a possible involvement of these neurons in the extraction of self-motion parameters from the opticflow,which might be useful,for instance,for stabilizing thefly’s head duringflight manoeuvres.A recent study[2]has shown that a simplified computational model of the tangential neu-rons as a weighted sum offlow measurements was able to reproduce the observed response fields.The weights were chosen according to an optimality principle which minimizes the output variance of the model caused by noise and distance variability between differ-ent scenes.The question on how the output of such processing units could be used for self-motion estimation was left open,however.In this paper,we want tofill a part of this gap by presenting a classical linear estimation approach that extends a special case of the previous model to the complete self-motion problem.We again use linear combinations of localflow measurements but,instead of prescribing afixed motion axis and minimizing the output variance,we require that the quadratic error in the estimated self-motion parameters be as small as possible.From this0306090120150180−75−45−15154575azimuth (deg.)e l e v a t i o n (d e g .)Figure 1:Mercator map of the response field of the neuron VS7.The orientation of each arrow gives the local preferred direction (LPD),and its length denotes the relative local motion sensitivity (LMS).VS7responds maximally to rotation around an axis at an azimuth of about 30◦and an elevation of about −15◦(after [1]).optimization principle,we derive weight sets that lead to motion sensitivities similar to those observed in tangential neurons.In contrast to the previous model,this approach also yields the preferred motion directions and the motion axes to which the neural models are tuned.We subject the obtained linear estimator to a rigorous real-world test on a gantry carrying an omnidirectional vision sensor.2Modeling fly tangential neurons as optimal linear estimators for self-motion2.1Sensor and neuron modelIn order to simplify the mathematical treatment,we assume that the N elementary motion detectors (EMDs)of our model eye are arranged on the unit sphere.The viewing direction of a particular EMD with index i is denoted by the radial unit vector d i .At each viewing direction,we define a local two-dimensional coordinate system on the sphere consisting of two orthogonal tangential unit vectors u i and v i (Fig.2a ).We assume that we measure the local flow component along both unit vectors subject to additive noise.Formally,this means that we obtain at each viewing direction two measurements x i and y i along u i and v i ,respectively,given byx i =p i ·u i +n x,iandy i =p i ·v i +n y,i ,(1)where n x,i and n y,i denote additive noise components and p i the local optic flow vector.When the spherical sensor translates with T while rotating with R about an axis through the origin,the self-motion-induced image flow p i at d i is [3]p i =−µi (T −(T ·d i )d i )−R ×d i .(2)µi is the inverse distance between the origin and the object seen in direction d i ,the so-called “nearness”.The entire collection of flow measurements x i and y i comprises theoptic flow vectors LPD unit vectorsLMSs summationxa.b.Figure 2:a.Sensor model:At each viewing direction d i ,there are two measurements x i and y i of the optic flow p i along two directions u i and v i on the unit sphere.b.Simplified model of a tangential neuron:The optic flow and the local noise signal are projected onto a unit vector field.The weighted projections are linearly integrated to give the estimator output.input to the simplified neural model of a tangential neuron which consists of a weighted sum of all local measurements (Fig.2b )ˆθ=N iw x,i x i +N iw y,i y i (3)with local weights w x,i and w y,i .In this model,the local motion sensitivity (LMS)isdefined as w i = (w x,i ,w y,i ) ,the local preferred motion direction (LPD)is parallel to thevector 1w i(w x,i ,w y,i ).The resulting LMSs and LPDs can be compared to measurements on real tangential neurons.As our basic hypothesis,we assume that the output of such model neurons is used to es-timate the self-motion of the sensor.Since the output is a scalar,we need in the simplest case an ensemble of six neurons to encode all six rotational and translational degrees of freedom.The local weights of each neuron are chosen to yield an optimal linear estimator for the respective self-motion component.2.2Prior knowledgeAn estimator for self-motion consisting of a linear combination of flow measurements nec-essarily has to neglect the dependence of the optic flow on the object distances.As a consequence,the estimator output will be different from scene to scene,depending on the current distance and noise characteristics.The best the estimator can do is to add up as many flow measurements as possible hoping that the individual distance deviations of the current scene from the average will cancel each other.Clearly,viewing directions with low distance variability and small noise content should receive a higher weight in this process.In this way,prior knowledge about the distance and noise statistics of the sensor and its environment can improve the reliability of the estimate.If the current nearness at viewing direction d i differs from the the average nearness ¯µi over all scenes by ∆µi ,the measurement x i can be written as (see Eqns.(1)and (2))x i =−(¯µi u i ,(u i ×d i )) T R +n x,i −∆µi u i T ,(4)where the last two terms vary from scene to scene,even when the sensor undergoes exactlythe same self-motion.To simplify the notation,we stack all2N measurements over the entire EMD array in the vector x=(x1,y1,x2,y2,...,x N,y N) .Similarly,the self-motion components along the x-,y-and z-directions of the global coordinate systems are combined in the vector θ=(T x,T y,T z,R x,R y,R z) ,the scene-dependent terms of Eq.(4)in the2N-vector n=(n x,1−∆µ1u1T,n y,1−∆µ1v1T,....) and the scene-independent terms in the 6xN-matrix F=((−¯µ1u 1,−(u1×d1) ),(−¯µ1v 1,−(v1×d1) ),....) .The entire ensemble of measurements over the sensor can thus be written asx=Fθ+n.(5) Assuming that T,n x,i,n y,i andµi are uncorrelated,the covariance matrix C of the scene-dependent measurement component n is given byC ij=C n,ij+Cµ,ij u i C T u j(6) with C n being the covariance of n,Cµofµand C T of T.These three covariance matrices, together with the average nearness¯µi,constitute the prior knowledge required for deriving the optimal estimator.2.3Optimized neural modelUsing the notation of Eq.(5),we write the linear estimator asˆθ=W x.(7) W denotes a2N x6weight matrix where each of the six rows corresponds to one model neuron(see Eq.(3))tuned to a different component ofθ.The optimal weight matrix is chosen to minimize the mean square error e of the estimator given bye=E( θ−ˆθ 2)=tr[W CW ](8) where E denotes the expectation.We additionally impose the constraint that the estimator should be unbiased for n=0,i.e.,ˆθ=θ.From Eqns.(5)and(7)we obtain the constraint equationW F=16x6.(9) The solution minimizing the associated Euler-Lagrange functional(Λis a6x6-matrix of Lagrange multipliers)J=tr[W CW ]+tr[Λ (16x6−W F)](10) can be found analytically and is given byW=12ΛF C−1(11)withΛ=2(F C−1F)−1.When computed for the typical inter-scene covariances of a flying animal,the resulting weight sets are able to reproduce the characteristics of the LMS and LPD distribution of the tangential neurons[2].Having shown the good correspondence between model neurons and measurement,the question remains whether the output of such an ensemble of neurons can be used for some real-world task.This is by no means evi-dent given the fact that-in contrast to most approaches in computer vision-the distance distribution of the current scene is completely ignored by the linear estimator.3Experiments3.1Linear estimator for an office robotAs our test scenario,we consider the situation of a mobile robot in an office environment. This scenario allows for measuring the typical motion patterns and the associated distance statistics which otherwise would be difficult to obtain for aflying agent.-75-45-15e l e v a t i o n (d e g .)a.Figure 3:Distance statistics of an indoor robot (0azimuth corresponds to forward direc-tion):a.Average distances from the origin in the visual field (N =26).Darker areas represent larger distances. b.Distance standard deviation in the visual field (N =26).Darker areas represent stronger deviations.The distance statistics were recorded using a rotating laser scanner.The 26measurement points were chosen along typical trajectories of a mobile robot while wandering around and avoiding obstacles in an office environment.The recorded distance statistics therefore reflect properties both of the environment and of the specific movement patterns of the robot.From these measurements,the average nearness ¯µi and its covariance C µwere computed (cf.Fig.3,we used distance instead of nearness for easier interpretation).The distance statistics show a pronounced anisotropy which can be attributed to three main causes:(1)Since the robot tries to turn away from the obstacles,the distance in front and behind the robot tends to be larger than on its sides (Fig.3a ).(2)The camera on the robot usually moves at a fixed height above ground on a flat surface.As a consequence,distance variation is particularly small at very low elevations (Fig.3b ).(3)The office environment also contains corridors.When the robot follows the corridor while avoiding obstacles,distance variations in the frontal region of the visual field are very large (Fig.3b ).The estimation of the translation covariance C T is straightforward since our robot can only translate in forward direction,i.e.along the z -axis.C T is therefore 0everywhere except the lower right diagonal entry which is the square of the average forward speed of the robot (here:0.3m/s).The EMD noise was assumed to be zero-mean,uncorrelated and uniform over the image,which results in a diagonal C n with identical entries.The noise standard306090120150180-75-45-15154575azimuth (deg.)e l e v a t i o n (d e g .)a.azimuth (deg.)e l e v a t i o n (d e g .)b.Figure 4:Model neurons computed as part of the linear estimator.Notation is identical to Fig.1.The depicted region of the visual field extends from −15◦to 180◦azimuth and from −75◦to 75◦elevation.The model neurons are tuned to a .forward translation,and b .to rotations about the vertical axis.deviation of 0.34deg./s was determined by presenting a series of natural images moving at 1.1deg./s to the flow algorithm used in the implementation of the estimator (see Sect.3.2).¯µ,C µ,C T and C n constitute the prior knowledge necessary for computing the estimator (Eqns.(6)and (11)).Examples of the optimal weight sets for the model neurons (corresponding to a row of W )are shown in Fig.4.The resulting model neurons show very similar characteristics to those observed in real tangential neurons,however,with specific adaptations to the indoor robot scenario.All model neurons have in common that image regions near the rotation or translation axis receive less weight.In these regions,the self-motion components to be esti-mated generate only small flow vectors which are easily corrupted by noise.Equation (11)predicts that the estimator will preferably sample in image regions with smaller distance variations.In our measurements,this is mainly the case at the ground around the robot (Fig.3).The rotation-selective model neurons weight image regions with larger distances more highly,since distance variations at large distances have a smaller effect.In our exam-ple,distances are largest in front and behind the robot so that the rotation-selective neurons assign the highest weights to these regions (Fig.3b ).3.2Gantry experimentsThe self-motion estimates from the model neuron ensemble were tested on a gantry with three translational and one rotational (yaw)degree of freedom.Since the gantry had a position accuracy below 1mm,the programmed position values were taken as ground truth for evaluating the estimator’s accuracy.As vision sensor,we used a camera mounted above a mirror with a circularly symmetric hyperbolic profile.This setup allowed for a 360◦horizontal field of view extending from 90◦below to 45◦above the horizon.Such a large field of view considerably improves the estimator’s performance since the individual distance deviations in the scene are more likely to be averaged out.More details about the omnidirectional camera can be found in [4].In each experiment,the camera was moved to 10different start positions in the lab with largely varying distance distributions.After recording an image of the scene at the start position,the gantry translated and rotated at various prescribed speeds and directions and took a second image.After the recorded image pairs (10for each type of movement)were unwarped,we computed the optic flow input for the model neurons using a standard gradient-based scheme [5].e s t i m a t o r r e s p o n s e [%]46810121416182022468101214161820true self-motione s t i m a t e d s e lf -m o t i o ntranslationrotationa.b.Figure 5:Gantry experiments:Results are given in arbitrary units,true rotation values are denoted by a dashed line,translation by a dash-dot line.Grey bars denote translation estimates,white bars rotation estimates a.Estimated vs.real self-motion;b.Estimates of the same self-motion at different locations;c.Estimates for constant rotation and varying translation;d.Estimates for constant translation and varying rotation.The average error of the rotation rate estimates over all trials (N=450)was 0.7◦/s (5.7%rel.error,Fig.5a ),the error in the estimated translation speeds (N=420)was 8.5mm/s (7.5%rel.error).The estimated rotation axis had an average error of magnitude 1.7◦,the estimated translation direction 4.5◦.The larger error of the translation estimates is mainly caused by the direct dependence of the translational flow on distance (see Eq.(2))whereas the rotation estimates are only indirectly affected by distance errors via the current translational flow component which is largely filtered out by the LPD arrangement.The larger sensitivity of the translation estimates can be seen by moving the sensor at the same translation and rotation speeds in various locations.The rotation estimates remain consis-tent over all locations whereas the translation estimates show a higher variance and also a location-dependent bias,e.g.,very close to laboratory walls (Fig.5b ).A second problem for translation estimation comes from the different properties of rotational and translational flow fields:Due to its distance dependence,the translational flow field shows a much wider range of values than a rotational flow field.The smaller translational flow vectors are often swamped by simultaneous rotation or noise,and the larger ones tend to be in the upper saturation range of the used optic flow algorithm.This can be demonstrated by simultane-ously translating and rotating the semsor.Again,rotation estimates remain consistent while translation estimates are strongly affected by rotation (Fig.5c and d ).4ConclusionOur experiments show that it is indeed possible to obtain useful self-motion estimates from an ensemble of linear model neurons.Although a linear approach necessarily has to ignore the distances of the currently perceived scene,an appropriate choice of local weights and a largefield of view are capable of reducing the influence of noise and the particular scene distances on the estimates.In particular,rotation estimates were highly accurate-in a range comparable to gyroscopic estimates-and consistent across different scenes and different simultaneous translations.Translation estimates,however,turned out to be less accurate and less robust against changing scenes and simultaneous rotation.The components of the estimator are simplified model neurons which have been shown to reproduce the essential receptivefield properties of thefly’s tangential neurons[2].Our study suggests that the output of such neurons could be directly used for self-motion esti-mation by simply combining them linearly at a later integration stage.As our experiments have shown,the achievable accuracy would probably be more than enough for head stabi-lization under closed loop conditions.Finally,we have to point out a basic limitation of the proposed theory:It assumes linear EMDs as input to the neurons(see Eq.(1)).The output offly EMDs,however,is only linear for very small image motions.It quickly saturates at a plateau value at higher image velocities.In this range,the tangential neuron can only indicate the presence and the sign of a particular self-motion component,not the current rotation or translation velocity.A linear combination of output signals,as in our model,is no more feasible but would require some form of population coding.In addition,a detailed comparison between the linear model and real neurons shows characteristic differences indicating that tangential neurons usually operate in the plateau range rather than in the linear range of the EMDs[2].As a conse-quence,our study can only give a hint on what might happen at small image velocities.The case of higher image velocities has to await further research.AcknowledgmentsThe gantry experiments were done at the Center of Visual Sciences in Canberra.The authors wish to thank J.Hill,M.Hofmann and M.V.Srinivasan for their help.Finan-cial support was provided by the Human Frontier Science Program and the Max-Planck-Gesellschaft.References[1]Krapp,H.G.,Hengstenberg,B.,&Hengstenberg,R.(1998).Dendritic structure and receptive field organization of optic low processing interneurons in thefly.J.of Neurophysiology,79,1902-1917.[2]Franz,M.O.&Krapp,H C.(2000).Wide-field,motion-sensitive neurons and matchedfilters for opticflowfields.Biol.Cybern.,83,185-197.[3]Koenderink,J.J.,&van Doorn,A.J.(1987).Facts on opticflow.Biol.Cybern.,56,247-254.[4]Chahl,J.S,&Srinivasan,M.V.(1997).Reflective surfaces for panoramic imaging.Applied Optics,36(31),8275-8285.[5]Srinivasan,M.V.(1994).An image-interpolation technique for the computation of opticflow and egomotion.Biol.Cybern.,71,401-415.。

Purity & Innovation3THE SAES GETTERS GROUP IS THE WORLD LEADER IN A VARIETY OF SCIENTIFIC AND INDUSTRIAL APPLICATIONS WHERE STRINGENT VACUUM CONDITIONS OR ULTRA-HIGH PURITY GASES ARE REQUIRED.BRINGING A WEALTH OF EXPERIENCE IN SPECIAL METALLURGY AND MATERIAL SCIENCE TO THE NEEDS OF THE ADVANCED MATERIAL INDUSTRY IS OUR NEW CHALLENGE FOR THE 21ST CENTURY.41947SAES, Società Apparecchi Elettrici e Scientifici based in Milan (Italy), beco-mes operative through investments byfamilies della Porta and Canale.1951Development of the first stablebarium-aluminum alloy.Pioneering the development of the getter technology, the SAES®Getters Group is the world leader in a variety of scientific and industrial applications where stringent vacuum conditions or ultra-high purity gases are required.For nearly sixty years, our getter solutions have been fostering and supporting technological innovation in the information display and lamp industries, in ultra-high vacuum systems, in a wide range of electronic device-based applications, and in the vacuum thermal insulation. The Group also delivers solutions for ultra-high purity gas handling to the semiconductor, fiber optics and otherhi-tech markets.Bringing a wealth of experience in special metallurgy and material science to the needs of the advanced material industry is our new challenge for the 21st century.We broaden our vision and keep on supporting your innovation. By nature.By leveraging the core competencies in special metallurgy, material science and thermal processes, the SAES Getters Group broadens its corporate vision and expands its business in the advanced material niche markets, with the introduction of:advanced optical crystals for the optoelectronic device and solid-state laser marketsshape memory alloys for components primarily used in the automotive, transportation and domotics industries metalorganic materials for chemical vapor deposition getters for MEMS/MOEMS and microelectronic hermetically packaged devicesAn outstanding Research & Development structure, based at the Group’s headquarters near Milan, Italy, is committed to technological excellence and keeps the Group at the forefront in innovation.A total production capacity distributed at 8 manufacturing plants spanning across 3 continents, a worldwide-based sales & service network, nearly 1,000 employees allow the Group to combine multicultural resources, skills and expertise to form a truly global enterprise, capable to best support customers around the world, 24 hours a day.SAES Getters has been listed on the Italian Stock Exchange Market, Star Segment, since 1986.Purity by Tradition, Innovation by Nature56Our VisionSAES Getters Group’s vision is to be the leading globalsupplier of advanced materials to niche markets characterized by high growth potential in the high-tech business segments.By leveraging our:sixty-year corporate culture of Research & Development and manufacturing excellenceworldwide leadership in the traditional getter-based application fieldscommitment to product and process qualitysound financial position and business integritythe SAES Getters Group is uniquely positioned to offer competitive advantages to the advanced material markets, as well as to support customers’ innovation and productivity enhancement through the provision oftechnological value.1958New headquarters and a firstmass-production plant are openedin Milan.1966Launch of Total Yield Flash Getters,extending color TV tube lifetime from 300 to 10,000 hours.Our Strategy SAES Getters Group’s business strategy builds on the following foundations:expanding our business portfolio and market leadership, through the development of new products within core competence areas pursuing the Group’s growth through selected acquisitions and strategic business partnerships fostering long-term business relationships with the ultimate goal to generate value for our customers, while enhancing the Group’s technological expertise and market competitiveness focusing on customer global support and satisfaction, maximizing quality and production efficiency through vertical integration, improved operational performance and attentive investment strategy committing to long-term success and stability, while practicing an ethic of responsibility and transparency towards stakeholders 7A Global Presence9Cologne GERMANY Daventry GREAT BRITAINCleveland OH – USA Colorado Springs CO – USA San Luis Obispo CA - USASAES Getters S.p.A.Group Headquarters Viale Italia 7720020 Lainate (Milan) - Italy Ph. +39 02 93178 1Fax +39 02 93178 320info @ 10Nanjing CHINASeoul KOREAMoscow RUSSIAShanghai CHINATokyo JAPANJincheon-kun KOREAAvezzano (AQ)ITALYSINGAPORE1112UNITED STATESSAES Getters USA, Inc.1122 East Cheyenne Mountain Blvd.Colorado Springs, CO 80906Ph. +1 719 576 3200 - Fax +1 719 576 5025susa@SAES Pure Gas, Inc.4175 Santa Fe RoadSan Luis Obispo, CA 93401Ph. +1 805 541 9299 - Fax +1 805 541 9399spg@EUROPESAES Getters S.p.A.Viale Italia 7720020 Lainate (Milan) - ItalyPh. +39 02 93178 1 - Fax +39 02 93178 320info@SAES Getters (Deutschland) GmbHGerolsteiner Strasse 1D 50937 Cologne - GermanyPh. +49 221 944 0750 - Fax +49 221 944 0754saesgermany@SAES Getters (GB) Ltd.Heritage House - Vicar LaneDaventry NN11 5AA - Great BritainPh. +44 1327 310777 - Fax +44 1327 310555saesgb@SAES Getters S.p.A. - Moscow Representative Office45, 24/2 Usievitch Street125315 Moscow - RussiaPh./Fax +7 095 7241228saesgetters@mtu-net.ruASIANanjing SAES Huadong Getters Co., Ltd.56 Xingangdadao, XinshengweiNanjing Economic & Technical Development ZoneNanjing 210038, Jiangsu Province - P.R. of ChinaPh. +86 25 8580 2335 - Fax +86 25 8580 1639saeschina@SAES Getters Technical Service Shanghai Co., Ltd.No. 415 Guo Shou Jing RoadZhangjiang Hi-Tech Park, Pudong New AreaShanghai 201203 - P.R. of ChinaPh. +86 21 5080 3434 - Fax +86 21 5080 3008saeschina@SAES Getters S.p.A. - Shanghai Representative OfficeNo. 415 Guo Shou Jing RoadZhangjiang Hi-Tech Park, Pudong New AreaShanghai 201203 - P.R. of ChinaPh. +86 21 5080 3434 - Fax +86 21 5080 3005saeschina@SAES Getters Singapore PTE, Ltd.6 Temasek BoulevardSuntec Tower Four #41-06Singapore 038986Ph. +65 6887 3343 - Fax +65 6887 3445saessingapore@SAES Getters Korea Corporation13th FL. Shinil B/D, 143-39Samsung-dong, Kangnam-kuSeoul 135-090 - KoreaPh. +82 2 3404 2400 - Fax +82 2 3452 4510www.saesgetters.co.krsaeskorea@SAES Getters Japan Co., Ltd.2nd Gotanda Fujikoshi Bldg.23-1 Higashi Gotanda 5-Chome, Shinagawa-kuTokyo 141 - JapanPh. +81 3 542 00431 - Fax +81 3 542 00438www.saesgetters.jpsaesjapan@Sales & Service Locations1970-79SAES Getters establishes sales subsidia-ries in Europe to fulfill the growing mar-ket demand.1973The new Japanese sales subsidiary is inaugurated in Tokyo.Manufacturing LocationsUNITED STATESSAES Getters USA, Inc.1122 East Cheyenne Mountain Blvd.Colorado Springs, CO 80906Ph. +1 719 576 3200 - Fax +1 719 576 50255604 Valley Belt RoadIndependence Cleveland, OH 44131Ph. +1 216 661 8488 - Fax +1 216 661 8796SAES Pure Gas, Inc.4175 Santa Fe RoadSan Luis Obispo, CA 93401Ph. +1 805 541 9299 - Fax +1 805 541 9399EUROPESAES Getters S.p.A.Viale Italia 7720020 Lainate (Milan) - ItalyPh. +39 02 93178 1 - Fax +39 02 93178 320SAES Advanced Technologies S.p.A. Nucleo Industriale67051 Avezzano (AQ) - ItalyPh. +39 0863 4951 - Fax +39 0863 495530ASIANanjing SAES Huadong Getters Co., Ltd.56 Xingangdadao, XinshengweiNanjing Economic & Technical Development Zone Nanjing 210038, Jiangsu Province - P .R. of China Ph. +86 25 8580 2335 - Fax +86 25 8580 1639SAES Getters Technical Service Shanghai Co., Ltd.No. 415 Guo Shou Jing RoadZhangjiang Hi-Tech Park, Pudong New Area Shanghai 201203 - P .R. of ChinaPh. +86 21 5080 3434 - Fax +86 21 5080 3008SAES Getters Korea Corporation 256-6, Okdong-ri, Ducksan-myun Jincheon-kun, Chungbuk-do - KoreaPh. +82 43 537 6000 - Fax +82 43 537 600813Research & Innovation1516THE HISTORY OF SAES GETTERS PROVES HOW INNOVATION AND HIGH QUALITY R&D FUEL ALL COMPANY’S ACTIVITIES. FOR SAES’ SCIENTISTS INNOVATION IS A MIND SET. THROUGH TECHNOLOGY PARTNERSHIPS AND ALLIANCES, THE SAES GETTERS GROUP HAS BEEN ABLE TO SUPPORT CUSTOMERS’ CUTTING-EDGE APPLICATIONS WITH THE DEVELOPMENT OF HIGHL Y INNOVATIVE SOLUTIONS FOR THE LAST 60 YEARS.1977SAES Getters opens SAES GettersU.S.A. Inc. at Colorado Springs, CO, itsfirst US sales and manufacturing facility.1981Development of low activation tempera-ture alloys for use in vacuum bottles and other vacuum insulated devices.The speed of the technological evolution that has characterized the last century and that is going toaccompany our lives in the new millennium requires a total commitment to innovation. The history of SAES Getters proves how innovation and high quality R&D fuel all company’s activities.For SAES’ scientists innovation is a mind set. They are challenged by customers’ requests, real market innovators,and are supported by a management policy that every year allocates approximately 8% of sales revenues to research and development activities.In the new facility in Lainate, in the Milan area, corporate laboratories cover an area of 3,300 square meters, where nearly 100 people are daily committed to investigate problems, develop proposals and test solutions using highly advanced instrumentation and mathematical modeling.Innovation at SAES is treated as a complex process that starts with the idea conception and ends with the launch of the related products onto the market.The innovation process takes place in the frame of our Research & Innovation function which includes, besides the Research department, also the Process Technology and Engineering departments.In order to better focus its Research & Innovation activities,the SAES Getters Group has recently adopted the Stage-Gate ®approach for the management of innovation projects, which enables a more effective utilization of resources through a structured screening of the new projects and of their development.The Group founds its business on competencies that have been acquired in sixty years of activity. The deep knowledge of special metallurgy and vacuum technology paved the way for broader competencies in material science and interests in advanced material applications.These, coupled with the understanding of the gas-surface interaction, and the know-how in UHP gas purification and analysis represent the core and the strength of SAES’capabilities.A deep understanding of the market needs guides SAES’R&I and the researchers and scientists highly value continuous relationships with industry key players. Not only is this essential to develop the right product with the required features: through technology partnerships and alliances, the SAES Getters Group has been able to support customers’ cutting-edge applications with the development of highly innovative solutions.The over 330 technologies patented during its entire business activity and the hundreds of scientific publications confirm SAES Getters’ continuous and unfalteringcommitment to innovation, as well as testify to a corporate policy firmly oriented to a strong intellectual property protection.At the Heart of SAES’ Excellence17Special MetallurgySAES’ skills in metallurgy stem from a scientific approach followed by the company in developing and characterizing getter alloys based on reactive metals. The skills include key aspects, such as the mastering of alloy melting techniques,the understanding of compositional and micro structural aspects of alloys, powder handling and characterization, as well as contamination related issues.Material ScienceThe deep understanding of special metallurgy and powder metallurgy developed into a more comprehensive and deeper knowledge in material science. This prompted SAES to expand its interests into new advanced materialdevelopments and applications, such as advanced optical crystals, shape memory alloys and metalorganic precursors for chemical vapour deposition (MOCVD).18The understanding of gas-surface interaction phenomena has guided SAES to advanced tailor-made getters and to catalytic materials for environmental applications.Getter Film DepositionThe increased miniaturization of flat displays and electronic devices called for thinner gettering solutions and SAES Getters has responded by offering porous getter filmsdeposited on different substrates through a proprietary sintering process.Our Core CompetenciesThe innovation process at the SAES Getters Group founds its solid basis on core competencies developed in a6-decade long activity, during which SAES Laboratories have become a center of excellence for material science and have developed an unparalleled know-how in ultra-high vacuum technology, gas-surface interaction and gas purification and analysis. These specific competencies, supported by strong technical capabilities in mechanical engineering, electronics and process simulation, allow the Group to offer premium solutions for highly technological markets. SAES’ productapplications span from the display industry, in all its forms from conventional cathode ray tubes to the new generations offlat displays, to the manufacturing of components used in several and diversified electronic devices ranging from x-ray tubes to wafer-level MEMS, from applications in particle accelerators and in large vacuum systems for physical experiments to solutions supporting the lamp, semiconductor, optoelectronic and automotive industries.1984The Avezzano industrial complex is esta-blished near Rome, for vertical integra-tion and quality control of all getter alloys used for the production of evaporable and non-evaporable getters.1987Acquisition of SAES Pure Gas Inc. in San Luis Obispo, California, a leading com-pany in the engineering and manufactu-ring of bulk and point-of-use gas purifiers for the semiconductor industry and other high-tech markets.Ultra-High Vacuum TechnologyCreating and maintaining ultra-high vacuum in a variety of systems pose challenges that only a deep knowledge of the various facets of vacuum science and technology can overcome. Thanks to its wealth of experience and to sixty years of leadership in vacuum related products, the SAES Getters Group can offer an all-round approach: the development of getter pumps adds up to the know-how built in the selection of materials and components, the evaluation of the outgassing and permeation and the vacuum system engineering.Gas-Surface InteractionThe deep understanding of the gas-surface phenomena has played a fundamental role in the optimization of the getter alloys and it has also allowed the development of special catalysts addressed to gas purification for the demanding semiconductor industry and to environmental applications.Chemical and Physical AnalysisA key point in understanding the material properties is their characterization via chemical and physical analysis. Thecapability to carry out the complete series of tests in-house guarantees SAES Getters’ on-time and highly reliable results. Among the wide range of tests available at SAES Labs are: Atomic Absorption Spectrophotometry,Inductively Coupled Plasma Spectrophotometry and Scanning Electron Microscopy.Gas PurificationThe knowledge of the kinetics of the gas-surface interaction and of the bulk diffusion mechanism of getter materials has been instrumental in the development of the gas purification business. SAES Getters researchers and engineers have a wealth of expertise in designing and assembling ultra-high purity systems, suitable to handle gases whose purity is better than 1 part-per-billion of atmospheric contaminants.Gas AnalysisSAES Labs have built up a long tradition of competencies in trace gas analyses, which started with the widespread use of various types of mass spectrometers to analyze the residual gases in vacuum devices. Residual gas analysis techniques have been specifically studied andimplemented not only for conventional vacuum systems,but also to cope with the increasing miniaturization of thevolumes to be analyzed, as in the case of OLED and MEMS devices.1920While it is widely recognized that the introduction of newproducts has an increasingly greater importance on sales aswell as on profits, specialized literature reports that onlyone in seven new projects delivers successful products.Thanks to its superior technology and to the sharpfocalization on customers’ future needs, the SAES GettersGroup has always enjoyed market success in thedevelopment and introduction of new products.To continue this tradition in a market environment which ischaracterized by increasing complexity and rapidtechnological changes, SAES Getters has recently updatedand implemented a more formalized process to select andrun new projects in the frame of the R&I function.The new product development (NPD) process is based onthe Stage-Gate methodology. This approach helps identify,organize, deploy and control the steps necessary to ensurethe effective and efficient implementation of new projects,with the ultimate goal to improve the chances for productsuccess and reduce time-to-market.Particularly, the innovation process articulates in a string ofphases of activities (stages) and of assessments anddecisions (gates). The project selection process is basedon criteria aimed at maximizing the chance of success bydeep assessment of a number of technical, economical andstrategic parameters.Key principles of SAES Getters’ adoption of the Stage-Gatemethodology are:production of superior, differentiated productscontinual interaction with customers and usersthroughout the development processsolid planning and sharp, early product definition beforedevelopment beginscross-functional effort and empowered project leadersand teamstough go/kill decisions at the gates to focus resources onthe most promising projectsfacilitation of clear project prioritization and resourceallocationThe Stage-Gate NPD process requires the contributionof different functions within the Group, from researchto engineering, from intellectual property protectionto marketing, supervised by a steering committee.A dedicated resource, the multi-project portfolio manager,coordinates the entire process and ensures it is evenly,correctly and timely carried out.The SAES Getters Group values the implementation of thisnew innovation management system as a great asset tocontinue the successful tradition in product innovation thathas distinguished the company throughout its entirebusiness history.Advancing the Management of Innovation1990First development of industrial purifiers(point-of-use, area and larger sizes),based on special getter alloys and cataly-tic materials, for purification of processgases to meet semiconductor manufac-turing requirements.1994Launch of high capacity non-evaporable getter pumps using newly developed sin-tered blades and disks, to improve mechanical characteristics and pumpingspeeds.1995Opening of the SAES Getters Shanghairepresentative office and of theSingapore sales subsidiary.In an increasingly competitive knowledge-based economy where intangible assets, such as brand awareness and innovation, have become keys to the corporate success, intellectual property protection plays a fundamental role. The SAES Getters Group has regarded its intellectual property as a strategic asset since its business inception and has always had an active role in managing the intellectual capital with dedicated resources.Key activities of SAES Intellectual Property department are the support during the patent approval process, as well as the obtainment of trademark registrations; the continuous monitoring of the new patents, released both in SAES’traditional application fields as well as in the new areas of business that the company is evaluating; and the technical assistance during litigation for an effective legal protection of the company’s rights.If the number of patents is an indication of the capability to innovate, SAES Getters can surely be proud of the over 330 inventions - including products and processes - that have been patented by the company in its sixty years of activity. The chart on the side, providing at-a-glance breakdown of the innovations patented for each business segment, shows how significantly the Group has contributed to a variety of application fields through the provision of innovative and competitive technological solutions. Intellectual Property: a Competitive AssetAlloys & GetteringDisplaysLampsVacuum InsulationGetter Pumps Gas Purification & Analysis Semiconductor TechnologyCatalysisOther12,12%27,58%5,76%10,61%5,45%19,09%3,33%2,42%13,64%21Quality, Environment, Safety and Ethics2324SAES GETTERS STRIVES TO PROVIDE CUSTOMERS WITH HI-TECH TOP QUALITY PRODUCTS AND SERVICES. THE GROUP’S FOCUS ON THE STRATEGIC VALUE OF TOTAL QUALITY MANAGEMENT ALSO RESULTS IN VERTICAL INTEGRATION OF THE PRODUCTION PROCESSES. THIS ALLOWS SAES GETTERS TO OPTIMIZE PRO-DUCT QUALITY WHILE FACILITATING THE DEVELOPMENT OF INNOVATIVE PRO-DUCTS AND PROCESSES WHICH ARE RESPECTFUL OF THE ENVIRONMENT AND OF THE NATURAL RESOURCES CONSERVATION.1996The Group’s headquarters and laborato-ries move to a new 15,000 sq. mt. faci-lity in Lainate, in the Milan area.1996Development of High Porosity Thin Film (HPTF) getters, for field emission dis-plays and for application in a wide range of electronic devices.To ensure continuous product improvement and strong market leadership, the SAES Getters Group remains loyal to its traditional core values and is committed to building an integrated Quality, Environment, Safety and Ethics Management System.The Group is certified under the UNI EN ISO9001:2000Quality Management System, with corporate certification covering all the manufacturing facilities.Quality excellence at SAES Getters results in vertical integration of production processes. This allows the Group to optimize product quality and to apply a close cost control policy, while facilitating the development of innovative products and processes, which are respectful of the environment and of natural resources conservation. The process approach is another distinctive mark of the Group’s Quality Management System. Structuring activities and related resources as a value-added process leads to improved performance, to consistent results and to the determination of clear responsibility in the managing of key activities.The evaluation of risks, consequences and impact of activities on stakeholders, as well as a more effective human resources management can also be more consistently achieved through process-structured activity.SAES Getters’ focus on the strategic value of total quality management shows in the continuous and attentive control of customer satisfaction, through adequate performance monitoring tools, and in the responsive implementation of proactive quality policies based on the analysis results.The SAES Getters Group recognizes that the protection of the global environment is an essential social duty for all human beings in the 21st century and that a substantial role needs to be played for the conservation of Earth resources.The Group is aware that all industrial activities, products and services may have a potential impact on theenvironment and, for this reason, strives to achieve a sound and consistent environmental performance for all its products and processes.All SAES Getters’ companies are committed to developing advanced products that exhibit safe andenvironment-friendly features by restricting the use of environmentally hazardous substances in products,encouraging a responsible exploitation of all natural resources and promoting projects for waste material recycling.To confirm SAES Getters’ efforts in preventing pollution and minimizing environment impacts, while still pursuing the continual improvement of their business performance, the following Group's manufacturing facilities have been ISO14001 certified: SAES Getters S.p.A. in Lainate (Milan),Italy; SAES Advanced Technologies S.p.A. in Avezzano, AQ,Italy; SAES Getters Korea Corporation in Jincheon-kun,Korea; and Nanjing SAES Huadong Getters Co. Ltd. in Nanjing, P .R. of China.We Care: an Integrated Management System 25Innovative Business Solutions2728SAES GETTERS’ PRODUCT LINES BASED ON THE EXPLOITATION OF THE PRO-PRIETARY GAS-SURFACE INTERACTION TECHNOLOGIES, TYPICALL Y USED IN VACUUM APPLICATIONS AND FOR ULTRA-HIGH PURITY GAS HANDLING, ARE OPE-RATED THROUGH TWO BUSINESS UNITS, THE INFORMATION DISPLAYS AND THE INDUSTRIAL APPLICATIONS UNITS.THE SIX BUSINESS AREAS UNDERL YING THIS STRUCTURE ARE COMMITTED TO ENSURING THE BEST CUSTOMER SATISFACTION IN SPECIFIC MARKET SEG-MENTS, WITH THE SUPPORT OF A SALES & SERVICE GLOBAL NETWORK DISTRI-BUTED ACROSS EUROPE, AMERICA AND ASIA.1998SAES Getters Technical Service Co. Ltd.is formed in Shanghai, China, and loca-ted in the Zhangjiang Hi-Tech Park,to best address the increasing technicalrequirements of the Chinese semicon-ductor and high-tech markets.1999Introduction of the St 787 series, non-evaporable getters capable of withstan-ding lamp manufacturing critical condi-tions and to deliver improved safety and ecological content.Information Displays Business Unit Cathode Ray TubesSAES Getters is the Color Cathode Ray Tube (CCRT) and Cathode Ray Tube (CRT) industry's worldwide leading provider of evaporable getters. With three manufacturing plants located in Italy, China and Korea exceeding an overall yearly production capacity of 250 million pieces, for the last 30 years SAES Getters has been unanimously recognized as the number one getter supplier for all kinds of cathode ray tubes. Excellent quality, extremely wide product portfolio, vertical integration and a worldwide logistic network are the key factors that motivate customers to choose SAES Getters not only as a supplier, but as a strategic business partner.Flat Panel DisplaysA strategic supplier of technology innovation for the flat panel display industry, SAES Getters offers solutions that meet the most severe requirements in terms of vacuum maintenance, impurity removal and controlled mercury release in cold cathode fluorescent lamps for LCDbacklighting. SAES’ evaporable and non-evaporable getters,desiccants, alkali metal sources and mercury dispensers support most of the cutting-edge flat panel developments –including Plasma displays, Field Emission Displays (FEDs),Vacuum Fluorescent Displays (VFDs) and Organic Light Emitting Diode (OLED) displays – by delivering the highest display efficiency and lifetime extension.2930Industrial Applications Business Unit2002Launch of IntegraTorr, an innovative non-evaporable getter (NEG) pumping solu-tion for particle accelerator vacuumchambers, developed by CERN and bran-ded by SAES Getters under a licenseagreement.LampsThe leading supplier of high quality, innovativenon-evaporable getters (NEG) and metal dispensing products, SAES Getters fully addresses and solves many key issues of the lamp industry related to vacuum purity maintenance, filling-gas purity and lamp processing for street, industrial and commercial lighting.SAES’ advanced mercury dispensers for fluorescent lamps support the most important lamp manufacturers worldwide in fully complying with stringent mercury dosing regulations, thus ensuring minimal environmental impact.Electronic DevicesThis Business Area offers solutions for an extremely wide and diversified range of electronic devices, whose functioning requires either vacuum or a high purity rare gas atmosphere to operate, and it addresses several different market segments: avionics, medical, security & defense, telecommunications and automotive. Applications of its sintered porous NEG products span from mature electron tubes to detectors and sensors, while the alkali metal dispenser product line fulfills the needs of the photosensitive device industry.。

Jason M. Shachat Curriculum Vita

Department of Economics National University of Singapore Block AS2, #06-05 1 Arts Link, Singapore 117570 ecsjs@nus.edu.sg Tel: (65) 6874 6019 Fax: (65) 6775 2646 Department of Business Policy NUS Business School Biz 1 Building, #03-26 1 Business Link, Singapore 117592 bizjs@nus.edu.sg Tel: (65) 6874 6273 Fax: (65) 6779 5059

Education University of Arizona, Tucson, AZ 1991-1996 Degrees: M.A. in Economics 1993, Ph.D. in Economics 1996 Thesis: “Heterogeneity and Equilibrium” Fields of Specialization: Experimental Economics, Microeconomic Theory, Econometrics

Tulane University, New Orleans, LA 1984-1988 Degree: B.S. in Mathematical Economics 1988

Professional Appointments Associate Professor, National University of Singapore joint appointment with the Departments of Economics and Business Policy, 2003-Present Research Staff Member, IBM TJ Watson Research Center, 2001– 2003 Assistant Professor of Economics, University of California, San Diego 1996–2001 Visiting Research Scientist, IBM TJ Watson Research Laboratories, Summer and Fall 2000 Visiting Research Scientist, HP Laboratories, Summer 1998 and 1999 Visiting Assistant Professor, GRID, Cachon, France October 1997

ZE/ZA 2-6 (1-4 bar(e)/14.5-58 psig) (30-522 kW/40-700 hp)LOW PRESSURE OIL -FREE AIR COMPRESSORSPROVIDING CONTINUOUS PRODUCTIVITYAT THE LOW EST OPERATIONAL COST As the cornerstone of many production processes, low pressure compressed air is essential tokeep the production going. Atlas Copco’s full range of low pressure oil-free air solutions offers a combination of high reliability and energy efficiency, providing a 100% certified supplyof oil-free air for a broad spectrum ofindustrial applications.Keeping your process up and running Especially in harsh and dusty environments, a reliable supply of compressed air is critical to ensure process continuity.Every ZE/ZA is designed, manufactured and tested to comply with ISO 9001 certification. T he totally enclosed IP55 motor is built to ensure continuous operation and exceptional reliability in dusty and humid environments.Driving down energy costsEnergy costs can amount to 80% of the Life Cycle Costs ofa compressor. T he generation of compressed air can account for more than 40% of a plant’s total electricity costs.Fully compliant with ISO 14001 standards, the ZE/ZA range helps to reduce costs: the IE3 / Nema premium efficiency motor and compression element with T eflon rotor coating and cooling jackets provide the highest air volume at the lowest energy consumption. T he integrated Variable Speed Drive (VSD) technology offers approximately 35% extra energy savingsby automatically tuning compressor flow to the requiredair demand.Protecting your reputationand productionIn virtually any application, oil contamination of the air supply causes serious productivity issues and increases costs. As the first manufacturer to receive ISO 8573-1 CLASS 0 (2010) certification for its oil-free air compressors, Atlas Copco has set a new standard in air purity. Focusing on the protectionof critical applications as well as today’s increasing quality demands, Atlas Copco offers TÜV-certified 100% oil-free air.Easy installationDelivered ready for use, ZE/ZA compressors come as all-in-one packages including a powerful controller and optional integrated aftercooler. T he complete scope of supply eliminates the need for extras and reduces installation to an absolute minimum, saving you time and money. Built for easy integration in your existing compressed air network, the ZE/ZAcompressors are up and running in no time.A COMPLETE PACKAGE FOR ALLYOUR APPLICATIONSBuilt to ensure complete product safety, the ZE/ZA compressors guarantee a continuous, highly reliable, energy-efficient and 100% oil-free air supply for decades in all your applications at the lowest possiblelife cycle cost.Pneumatic conveying – dense phase• Lowest energy cost, representing up to 80% of thecompressor life cycle cost.• Minimized downtime and maintenance cost thanks to innovativesingle stage screw compressor technology.Glass blowing• Higher pressure ratio for mold cooling up to 4 bar(e)/58 psig.• 100% Class 0 certified oil-free air resulting in the highestair quality.• Low energy consumption required for continuous operation.Fermentation• Lowest energy cost, representing up to 80% of the compressorlife cycle cost.• Low downtime and low maintenance cost thanks to innovativescrew compressor technology.• Very wide flow and pressure operational range.Mining• Increased energy efficiency and productivity with lowenvironmental impact.• Minimized downtime and maintenance cost thanks to innovativesingle stage screw compressor technology.CLASS 0: THE INDUSTRY STANDARD Oil-free air is used in all kinds of industries whereair quality is paramount for the end product and production process. T hese applications includefood and beverage processing, pharmaceutical manufacturing and packaging, chemical and petrochemical processing, semiconductor and electronics manufacturing, the medical sector, automotive paint spraying, textile manufacturingand many more. In these critical environments, contamination by even the smallest quantities of oil can result in costly production downtime and product spoilage.First in oil-free air technologyOver the past sixty years Atlas Copco has pioneered the development of oil-free air technology, resulting in a range of air compressors and blowers that provide 100% pure, clean air. Through continuous research and development, Atlas Copco achieved a new milestone, setting the standard for air purity as the first manufacturer to be awarded ISO 8573-1 CLASS 0 certification. Eliminating any riskAs the industry leader committed to meeting the needs of the most demanding customers, Atlas Copco requested the renownedTÜV institute to type-test its range of oil-free compressors and blowers. Using the most rigorous testing methodologies available, all possible oil forms were measured across a range of temperatures and pressures. T he TÜV found no traces of oil at allin the output air stream. T hus Atlas Copco is not only the first compressor and blower manufacturer to receive CLASS 0certification, but also exceeds ISO 8573-1 CLASS 0 specifications.concentration in total oil content).Oil pump• Oil pump mounted on thedrive shaft to reduce thenumber of auxiliary motors.• Reliable lubrication in wideoperation range.High precision drive system• Minimized transmission losses, noise andvibration levels.• Prolonged element lifetime thanks to the AGMAQ13/DIN Class 5 gears in the main drive.Coated rotors• Unique T eflon coating results in increasedefficiency, longer lifetime and protectionagainst corrosion.• High temperature resistance allowsoperating pressures up to4 bar(e)/58 psig.• The carbon steel rotors are synchronizedwith Nickel alloyed gears.Electrical cabinet• The standard rating of the short circuit currentprotection of the electrical cubicles is 50 kA (IEC),resp. 65 kA (CSA/UL).• Fixed speed and Variable Speed Drive.PRE-ENGINEERED SOLUTIONS FORALL YOUR NEEDSIn order to provide customers with a more flexible offering for requirements outside the standard product configuration, pre-engineered solutions are defined.Atlas Copco recognizes the need to combine the advantage of serial produced compressors with the typical requirements of the applications for this kind of equipment. T he specifications for low pressure compressors frequently require outdoor installation, operation in remote locations, often exposed to heavy duty conditions. Atlas Copco is offering pre-engineered kits to simplify the sales process.The special requests from original equipmentmanufacturers (OEM’s) for detailed documentation andmaterial certificates are provided through a simplifiedorder routine.Deviations on the standard motor selection (request fordifferent brand, oversized motor or motor options) andtests witnessed by the customer are other services whichare supported through the competent organizationbehind pre-engineered solutions.State-of-the-art screw compression element• Cooling jackets improve reliability and efficiency by ensuring rotor clearances are always kept to the absolute minimum.• Efficient shaft sealing eliminates the risk of oil leakage, reduces wear, and guarantees 100% oil-free air.Air in- and outtake• Air intakes and cooling air outlets are provided with mounting positions to allow easy ducting. • All gratings are provided with internal baffling to reduce the noise level.• Cooling air flows are internally separated to avoid recirculation.Integrated Variable Speed Drive (VSD) (optional)• Electrical cabinet with fully integrated frequency converter and control panel: no additional engineering and installation required.• Specifically selected drive components: the settings are fine-tuned to achieve maximum efficiency.• No blow-off of compressed air to the atmosphere at partial air flow requirement.• Optimized component selection.• Reduced installation cost.• No interferences.• EMC compliance tested and certified.Integrated air-cooled aftercooler*• The highly efficient cooling reduces energy consumption and dryer loads.• Variable speed fan motors allow for constant temperature control, energy savings and noise reduction (ZE 3-4).• Combination of stainless steel pre-cooler and aluminum aftercooler to cope with high heat stress and guarantee long life time.• User-friendliness is increased and costs reduced thanks to easy installation and easy access for cleaning.* Option.T otally enclosed motor• IP55 T EFC protection against dust and humidity.• Highly efficient motors according to IE3 (equal to NEMA Premium).• Dry motor coupling requires no lubrication, eliminating service requirements.Air filter• Quality air inlet filters provide high filtration class, process reliability and energy efficiency. • Long life time extends the service interval.Sound enclosure• Highly efficient noise reduction achieved through sound reflection in the sheet metal and noise absorption by the silencing foam. • Reduced noise installation cost of the compressor room.• Doors allow easy and quick access to all components.Load/unload regulation• Throttle valve controlled without the need for an external air supply.• No air compression at unload operation to reduce power consumption.Advanced element bearings• The bearings adapt easily to changing loads, providing the flexibility and efficiency to make production processes run smoothly.• Proven durability: two axial bearings limit internal leakage losses by maintaining small clearances between the rotor surfaces.Stainless-steel water-cooled aftercooler• Corrosion-resistant stainless steel tubes.• The risk of leaks is eliminated thanks to highly precise robot welding.• Cooling water outside tubes guided by baffles:- Low pressure drop.- No dead zones – limited fouling.- No degradation in cooler performance.- Easy cleaning.- Very long service intervals.StandardNEMA 4electric cubicleAdvanced Elektronikon ®unit controller• One integrated control system for compressor.• Overall system performance status with pro-activeservice indications, alarms for mailfunctions andsafety shutdowns.• Multi-language selectable user interface.• Designed to interface with the Atlas CopcoES central controller.• Remote control and monitoring is possible viaProfibus and Modbus communication.MotorsLow and medium voltage motors availablewith or without starter.VSD: DRIVING DOWN ENERGY COSTSOver 80% of a compressor’s lifecycle cost is taken up by the energy it consumes. Moreover, the generation of compressed air can account for more than 40% of a plant’s total electricity bill. T o cut your energy costs, Atlas Copco pioneered Variable Speed Drive (VSD) technology in the compressed air industry. VSD leads to major energy savings, while protecting the environment for future generations. T hanks to continual investments in this technology, Atlas Copco offers the widest range of integrated VSD compressors on the market.What is VSD technology?• In almost every production environment, air demand fluctuates depending on different factors (time of the day, week or even month).• Extensive measurements and studies of compressed air demand profiles show that many blowers have substantial variations in air demand. Only 8% of all installations have a more stable air demand. T ests prove that, even in this case, VSD compressors save energy.Varying air demand in 92% of all installationsIn almost every production environment, air demand fluctuates depending on different factors (time of the day, week or even month). Extensive measurements and studies of compressed air demand profiles show that 92% of all compressor and blower installations have substantial variations in air demand. Only 8% of all installations have a more stable air demand. T ests prove that, even in this case, VSD compressors save energy.EnergyEnergy savings with VSD Investment MaintenanceEnergy savings on average up to 35%Atlas Copco's VSD technology closely follows the air demand byautomatically adjusting the motor speed. T his results in large energysavings of up to 35%. T he Life Cycle Cost of a compressor can be cutby an average of 22%. In addition, lowered system pressure with VSDminimizes energy use across your production dramatically.T otal compressor lifecycle cost• 64% of all installations• Factory working 24 hrs/day:low demand at night & highdemand during the day • 28% of all installations • Factory working 2 shifts/day, no weekend work: erratically varying air demand • 8% of all installations • Factory working 2 shifts/day, no weekend work: typical 'fixed' speed applicationProfile 1M A X Profile 2M A X Profile 3M A XA STEP AHEAD IN MONITORING AND CONTROLSThe Elektronikon ® operating system offers a wide variety of control and monitoring features that allow you to increase your compressor’s efficiency and reliability. T o maximize energy efficiency, the Elektronikon ® controls the main drive motor and regulates system pressure within a predefined and narrow pressure band.• Improved user-friendliness: 5.7” color display with clearpictograms for easy readout.• Monitoring of running conditions and graphical indicationof the service plan.• Regulates system pressure within a predefined narrow pressure band.• Integrated energy savings functions like dual pressureset point, 4 different programmable week schedules.• Comprehensive icon indications and intuitive navigation.• 31 different languages including character-based languages.• Durable keyboard to resist tough treatment indemanding environments.• Internet-based compressor visualization using a simple Ethernet connection.• Remote control and advanced connectivity functions.Built-in intelligenceOnline & mobile monitoringMonitor your compressors over the Ethernet withthe new Elektronikon ® controller. Monitoringfeatures include warning indications, compressorshut-down and maintenance scheduling.An Atlas Copco App is available for iPhone/Androidphones as well as iPad and Android tablets. It allowsfingertip monitoring of your compressed air systemthrough your own secured network.• A remote monitoring system that helps you optimize yourcompressed air system and save you energy and cost.• It offers you a complete insight in your compressed air networkand anticipates on potential problems by warning you up-front.* Please contact your local sales representative for more information.SMART Link *:Data Monitoring ProgramOPTIMIZE YOUR SYSTEMWith the ZE/ZA, Atlas Copco provides an all-in-one standard package incorporating the latest technology in a built-to-last design. T o further optimize your ZE/ZA’s performance or to simply tailor it to your specific production environment, optional features are available.Standard scope of supply*For more details please contact your Atlas Copco sales representative.OptionsProduct range4321500100015002000(l/s)1800360052007200(m 3/h)ZE/A 2ZE/A 3-4ZA 5-62935 0570 14 © 2016, A t l a s C o p c o A i r p o w e r N V , B e l g i u m . A l l r i g h t s r e s e r v e d . D e s i g n s COMMITTED TO SUSTAINABLE PRODUCTIVITY We stand by our responsibilities towards our customers, towards the environment and the people around us. We make performance stand the test of time. T his is what we call – Sustainable Productivity.。

Low pressure oil-free rotary screw compressorsZE 3S37-90 kW / 50-120 hpProviding ruggedperformance at thelowest operationalcostAs in many production processes, low pressure compressed air isessential to keep the production going. Atlas Copco's NEW dedicatedrange of low pressure compressors offers rugged performance andprovides 100% certified oil-free air for a broad spectrum of pneumaticconveying industrial applications.ZE 3S37-90 kW / 50-120 hp – insert tablesProtecting your reputation andproductionIn virtually any application, oil contamination of the air supplycauses serious productivity issues and increased costs. Being thefirst manufacturer to receive ISO 8573-1 CLASS 0 (2010)certification for its oil-free air compressors, Atlas Copco has set astandard in air purity. Focusing on the protection of criticalapplications as well as today's increasing quality demands, AtlasCopco offers TUV-certified 100% oil-free air.Keeping your process up and runningEspecially in a harsh and dusty environment, a reliable supply ofcompressed air is critical to ensure process continuity. Everycompressor is desgined, manufactured and tested to comply withISO9001 certification. The latest innovation of screw elementdesign, the robust gearbox and totally enclosed IP55 motor isbuilt to ensure continuous operation and outstanding reliabilityeven in the dustiest, hottest and most humid environment.Easy installation and versatilityWe can deliver the compressor according to your need. Do youwant to have your own starter? No problem. Do you need aready-to-use-machine? Sure. All as standard. Designed and builtfor easy integration in your existing compressed air network or asnew installations.Assuring your peace of mindThrough continuous investment in our competent, committed andefficient service organization, Atlas Copco ensures superiorcustomer value by maximizing productivity. With a presence inover 170 countries, we offer professional and timely servicethrough interaction and involvement. Uptime is guaranteed bydedicated technicians and 24/7 availability.3•Lowest energy cost, representing up to 80% of the compressor life cycle cost.•Minimized downtime and maintenance cost thanks to innovative single stage screw compressor technology.Pneumatic conveying•On site compressor with higher unloading capacity than a truck mounted blower •Low noise emission•No risk for product contamination•Easy connection and faster unloading compared to truck unloaded installationsBulk transportationZE 3S 37-90 kW / 50-120 hp – insert tables•Higher pressure ratio for mold cooling•100% Class 0 certified oil-free air resulting in the highest air quality•Low energy consumption required for continuous operationGlass•Lowest energy cost, representing up to 80% of the compressor life cycle cost•24/7 uniterrupted pollution control thanks to proven reliable designFlue gas desulphurization5Oil-free air is used in all kinds of industries where air quality is paramount for the end product and production process. These applications include food and beverage processing, pharmaceutical manufacturing and packaging, chemical and petrochemical processing, semiconductor and electronics manufacturing, the medical sector, automotive paint spraying, textile manufacturing and many more. In these critical environments, contamination by even the smallest quantities of oil can result in costly production downtime and product spoilage.Class 0: Oil-free airOver the past sixty years Atlas Copco has pioneered the development of oil free air technology, resulting in a range of air compressors and blowers that provide 100% pure, clean air. With our CLASS 0 products, no oil is added during the compression process, and thus provides you with 100% pure, clean air when the atmosphere doesn't contain any oil particles. Through continuous research and development, Atlas Copco achieved a new milestone,setting the standard for air purity as the first manufacturer to be awarded ISO 8573-1 CLASS 0 certification.First in oil-free air technologyAs the industry leader committed to meeting the needs of the most demanding customers, Atlas Copco requested the renowned TÜV institute to type-test its range of oil-free compressors and ing the most rigorous testing methodologies available, all possible oil forms were measured across a range of temperatures and pressures. The TÜV found no traces of oil at all in the output air stream. Thus Atlas Copco is not only the first compressor and blower manufacturer to receive CLASS 0 certification, but also exceeds ISO 8573-1 CLASS 0 specifications.Eliminating any riskZE 3S 37-90 kW / 50-120 hp – insert tables•Inlet filter is combined with a silencer to reduce noise level andprotect the compression stage.•Handy pressure drop indication on the control panel.•Long lifetime extends the service interval.•Cooling jackets improve reliability and efficiency by ensuring rotor clearances always to the absolute minimum•Efficient shaft sealing eliminates the risk of oil leakage, reduces wear and guarantees 100% oil-free air•High performance coated screw rotors for increased efficiency, longer lifetime and protection against corrosion •Minimized transmission losses, noise and vibration levels •Prolonged element lifetime thanks to AGMA Q13/DIN Class 5 gears in the main drive•Directly driven by the gearbox•Oil injection nozzles spray the optimal amount anf temperature of filtered oil to each bearing/gear•Discharge pulsation damper attenuates dynamic pulsation levels in the air flow to the minimum1. Process air filter2. Equiped with the best Atlas Copco oil-free screw element3. High precision gearbox drivesystem4. Integrated oil pump5. Silencer7•IP55 TEFC protection against dust and humidity •Highly efficient motors according to IE3/NEMA Premium •Dry motor coupling requires no lubrication and eliminates service requirements•Optimal cooling in wide temperature operation range •Easy and quick cleaning possibility in harsh environment•Available for built in YD starter and several options •One integrated control system for compressor•Overall system performance status with pro-active service indications, alarms and safety shutdowns•Remote control and monitoring is possible with Profibus, Modbus and TCP/IP•Several built-in energy saving algorithms•Multi-language•Inlet baffle silencing with minimum pressure drop and high sound absorption characteristics•Sealed canopy panels and doors6. High efficient totally enclosed motor7. Oil cooler8. Electrical cubicle 9. Advanced Elektronikon®unit controller10. Silent canopyZE 3S37-90 kW / 50-120 hp – insert tablesInstallation flexibilityY ou will really like to install and work with our ZE 3S compressor, designed to fit into your process wherever you want.Installation flexibilityForget the difficulites of the old machines replacement. We make your life easier,replacing the outdated technology with our latest technology machines and with the smallest footprint. Building new installations with very small footprint makes you save space and thus investment cost.The smallest footprint in the flow and pressure rangeNo need for a dedicated compressor room, no need for excessively long piping… Y ou can install the ZE 3Scompressor wherever you want to use the bolt-on outdoor-kit.Outdoor operation9Atlas Copco's VSD technology closely follows the air demand by automatically adjusting the motor speed. This results in large energy savings of up to 35%. The Life Cycle Cost of a compressor can be cut by an average of 22%. In addition, lowered system pressure with VSD minimizes energy use across your production dramatically.What is unique about the integrated Atlas Copco VSD?•The Elektronikon®controls both the compressor and the integrated converter, ensuring maximum machine safety within parameters.•Flexible pressure selection with VSD reduces electricity costs.•Specific converter and motor design (with protected bearings) for the highest efficiency across the speed range.•Electric motor specifically designed for low operating speeds with clear attention to motor cooling and compressor cooling requirements.•All Atlas Copco VSD compressors are EMC tested and certified. Compressor operation does not influence external sources and vice versa.•Mechanical enhancements ensure that all components operate below critical vibration levels throughout the entire compressor speed range.•A highly efficient frequency converter in a cubicle ensures stable operation in high ambient temperatures up to 50°C/122°F•No ‘speed windows’ that can jeopardize the energy savings and the stable net pressure. Turndown capability of the compressor is maximized to 70-75%.•Net pressure band is maintained within 0.10 bar, 1.5 psi.Energy savings up to 35% ZE 3S37-90 kW / 50-120 hp – insert tables•Air intake with noise attenuating baffle system.•Air is filtered prior to entering the oil-free screw compressor element.•Internal compression in the oil-free screw element.•Discharge silencer reduces the pressure pulsation levels to the minimum.•Safety valve to protect the unit from over pressure.•Check valve is protecting backflow from the pressure network.•Air delivery to the system.Process flow•Integrated oil pump, mounted on the gearbox hence directly driven.•Oil suction from carter, integrated in the gearbox.•Bypass valve decides the exact amount of oil flow that is required for bearing- and gear cooling and lubrication.•Than oil is first pumped through the oil cooler.•Filtered cool oil is distributed to individually tuned oil nozzles per bearing and/or gear in oil-free screw compressor element and gearbox and in the cooling jacket of the element.•Internal drains recover all oil in the carter (in the gearbox).Oil flow•One cooling fan pulls fresh air from the back side of the unit.•That fresh air is pushed through the oil cooler, taking away the heat of the oil.•In parallel, the motor cooling fan also pulls fresh air from the backside of the unit. The motor fan-cowl ensures that air flows over the motor cooling fins.•The cubicle is cooled with fresh air taken-in from the atmosphere through filters in the front door.•Cubicle fans push the hot air out of the cubicle, in the canopy.•The hot canopy air (oil cooling heat, motor cooling heat and cubicle heat) can leave the canopy through a roof-top grating. A noise attenuating baffle is installed.Cooling flowThe Elektronikon ®unit controller is specially designed to maximize the performance of your compressors and air treatment equipment under a variety of conditions. Our solutions provide you with key benefits such as increased energy efficiency, lower energy consumption, reduced maintenance times and less stress… less stress for both you and your entire air system.Elektronikon ®MK5 TouchThe full color touch display gives you an easy-to-understand readout of the equipment’s running conditions.•Clear icons and intuitive navigation provides you fast access to all of the important settings and data.•Monitoring of the equipment running conditions andmaintenance status; bringing this information to your attention when needed.•Operation of the equipment to deliver reliable compressed air specified to your compressed air needs.•Built- in remote control and notification functions provided as standard, including simple- to-use integrated webpage.•Integrated SMART LINK•Built- in remote control and notifications functions provided as standard, including simple-to-use integrated webpage.•Support for 31 different languages, including character based languages.Intelligence is part of the packageMonitor your machines over the ethernet with the Elektronikon ®unit controller and the SMART LINK service. Monitoring features include warning indications, compressor shut-down, sensor trending and maintenance scheduling.Online & mobile monitoringDual set-point and automatic stopMost production processes create fluctuating levels of demand which, in turn, can create energy waste in low use periods. By using the Elektronikon®controller, you can manually or automatically switch between two different setpoints to optimize energy use and reduce costs at low usetimes. In addition, the sophisticated algorithm runs the drive motor only when needed. As the desired setpoint is maintained while the drive motor’s running time is minimized, energy consumption is kept to a minimum.SMARTLINKMonitor your compressed air installation with SMARTLINKKnowing the status of your compressed air equipment at all times is the surest way to achieve optimal efficiency and maximum availability.Go for energy efficiencyCustomized reports on the energy efficiency of your compressor room.Increase uptimeAll components are replaced on time, ensuring maximum uptime.Save moneyEarly warnings avoid breakdowns and production loss.Evolving towards compressed air managementSMARTLINK ServiceA mouse-click reveals the online service log. Get quotes for parts and additional service quickly and easily.SMARTLINK UptimeSMARTLINK Uptime additionally sends you an e-mail or text message whenever a warning requires your attention.SMARTLINK EnergySMARTLINK Energy gives you customized reports on the energy efficiency of your compressor room, in compliance with ISO 50001.•Save costs - Optimal maintenance will reduce the operational cost of your compressor system.•Increase operational efficiency - Our maintenance expertise makes your life easier when it comes to resource management.•High uptime and performance - Specialist service keeps your equipment running and protect your investment.Reduce your total cost of ownership and benefit fromoptimal performanceGenuine Parts, designed and produced to the exact specifications of your compressors, delivered right where and when you need them.•All parts, one package - Always have the needed part for your service intervention at hand.•Save money - A Service Kit costs less than the sum of its components if ordered separately•Less administration - Every Service Kit has a single part number,allowing you to create a simple purchase order which can be easily followed-up.Compressor parts at your doorstep:our Parts PlanAvoid financial surprises. Our Fixed Price Services combine the expertise of factory-trained technicians with the quality of our genuine compressor parts.•The best compressor parts - The unrivalled quality of our genuine parts results in optimal uptime, energy consumption and reliability.•An expert maintenance plan - Rely on the expertise of factory-trained Atlas Copco technicians.•Clear and easy - Tailored to your installation, site conditions, and production planning, every Fixed Price Service has a clear scope and price.Fixed Price Services: best compressorparts & maintenanceRely on trained Atlas Copco technicians and the unrivalled quality of our genuine parts.•Service reports - We help you achieving maximum energyefficiency by keeping you up to date of the status of your system.•Prevent breakdown - If our technicians spot an additional developing problem, they will propose a solution.•Top-priority emergency call out system - If an urgent repair is needed, you get priority assistance.Preventive Maintenance Plan for optimal compressor uptimeWe take care of all your compressor maintenance, upgrades,repairs and even breakdowns at an all-inclusive price.•Complete compressor care - On-time maintenance by expert service engineers, genuine parts, proactive upgrades and compressor overhauls.•Total risk coverage - This means we take care of all yourcompressor repairs and even breakdowns, without extra charges.•Ultimate efficiency - Fitting the latest drive line components gives you high standards of compressor efficiency and reliability.Complete compressor care with our Total Responsibility PlanZE 3S VSD (No-Starter)331.521.8995585470275771.7525.491554047027577229845495470275772.2532.678046047027577ZE 3S VSD 37 kW No-Starter2.536.37204254652757737501.521.81180695470275771.7525.41090645470275772291010595470275772.2532.694055547027577ZE 3S VSD 45 kW No-Starter2.536.3875515465275774560135030001.521.81385815470275781.7525.41295765470275782291205710470275782.2532.6113066547027578ZE 3S VSD 55 kW No-Starter2.536.310556204652757855751.7525.41650970470275782291560920470275782.2532.6147586547027578ZE 3S VSD 75 kW No-Starter2.536.31370805465275781.521.8164096546527578751001.7525.41640965465275782291640965465275782.2532.6164096546527578ZE 3S VSD 90 kW No-Starter2.536.316409654652757890120160035251500/1850 x 1250 x 172059/73 x 49x 68ZE 3S - 50 Hz (Plug & Play and No-starter)31.521.8955560771.7525.486050577229.063037077229.070041077229.0775455772.2532.6625370772.2532.6700410772.2532.6770455772.536.362537077ZE 3S 37 kW2.536.36954107737501.521.81065625771.7525.4106062577229.0950560772.2532.685550577 ZE 3S 45 kW2.536.3855500774560135030001.521.81175695781.7525.4117569078229.01170690782.2532.6105562078ZE 3S 55 kW2.536.310506207855751.521.81450850781.7525.4144585078229.01440845782.2532.6143584578ZE 3S 75 kW2.536.3128075578751001.521.81595940781.7525.4159093578229.01585930782.2532.6158093078 ZE 3S 90 kW2.536.315759307890120160035251500/1850 x 1250 x 172059/73 x 49 x 68ZE 3S - 60 Hz (Plug & Play and No-starter)31.521.8950560771.7525.486050577229.077045577ZE 3S 37 kW2.2532.67704557737501.521.81160680771.7525.4104561577229.0945555772.2532.685550077 ZE 3S 45 kW2.536.3850500774560135030001.521.81285755781.7525.4128075578229.01150680782.2532.6104061078ZE 3S 55 kW2.536.310356107855751.521.81560920781.7525.4155591578229.01410830782.2532.6141083078 ZE 3S 75 kW2.536.3127075078229.01550915782.2532.6154591078 ZE 3S 90 kW2.536.315459107890120160035251500/1850 x 1250 x 172059/73 x 49 x 68(1)Unit performance measured according to ISO 1217, Annex C & E, Edition 4 (2009)Reference conditions:- Absolute inlet pressure 1 bar (14.5 psi).- Intake air temperature 20°C (68°F).(2)A-weighted emission sound pressure level at the work station, Lp WSA (re 20 μPa) dB (with uncertainty 3 dB). Values determined according to noise level test code ISO 2151 and noise measurement standard ISO 9614.(3)L' = length of the unit including motor-backpackZE 3S - 50Hz (without motor)31.521.81450850-751001.521.81595940-1.7525.41590935-2291585930-2.2532.61580930-ZE 3S 37 kW2.536.31575930-90120110024251500/1850 x 1250 x 172059/73 x 49 x 68ZE 3S - 60Hz (without motor)31.521.81560920-1.7525.41555915-75100 229.01550915-2.2532.61545910-ZE 3S 37 kW2.536.31545910-90120110024251500/1850 x 1250 x 172059/73 x 49 x 68(4)Unit performance measured according to ISO 1217, Annex C & E, Edition 4 (2009) Reference conditions:- Absolute inlet pressure 1 bar (14.5 psi).- Intake air temperature 20°C (68°F).(5)L' = length of the unit including motor-backpackAir inlet filter✓Coated screw element with cooling jacket✓Air circuitCheck valve✓Discharge pulsation damper✓Outlet air flange DN125 DIN&ANSI✓Supplied oil-filled✓Completely pre-piped oil circuit✓Integrated oil pump✓Oil circuitOil cooler✓Oil filter✓Built-in oil breather system✓Motor IE3/NEMA3 induction motor, TEFC IP55✓No starter included✓Elektronikon® Touch controller with SMARTLINK✓CubicleSensors air & oil pressure & temperature✓LAN or Internet control/monitoring✓Sound attenuating canopy✓BodyworkFrame with forklift/pallet-jack slots✓Mechanical approval ASME or CE approval✓Blow-down valve✓VSD enabled motor✓YD starter✓Test certificate✓Witness test✓Wooden case packing✓Full option motor (anti-condensation heaters & PT1000's)✓Oversized motor✓No motor✓Outdoor canopy✓Separate air intake✓Heavy duty inlet filter✓SPM monitoring✓Winterization✓COMMITTED TO SUSTAINABLE PRODUCTIVITY We stand by our responsibilities towards our customers, towards theenvironment and the people around us. We make performance standthe test of time. This is what we call – Sustainable Productivity.。

第2期(总第388期)2024年2月商㊀业㊀经㊀济㊀与㊀管㊀理JOURNAL OF BUSINESS ECONOMICSNo.2(General No.388)Feb.2024收稿日期:2023-11-20基金项目:国家自然科学基金面上项目 创业型领导㊁创新行为与新创企业成长:基于中国高科技企业的跨层研究 (71972074)作者简介:易凌峰,男,教授,博士生导师,教育学博士,主要从事人力资源管理与知识管理研究;余志远(通讯作者),男,博士研究生,主要从事组织行为学与人力资源管理研究;陈浩宇,男,博士研究生,主要从事组织行为学与知识管理研究㊂提振士气还是制造焦虑?数字化转型对员工敬业度的双刃剑效应易凌峰,余志远,陈浩宇(华东师范大学经济与管理学院,上海200062)摘㊀要:文章从资源保存理论出发,构筑数字化转型对员工敬业度的影响模型,分析工作成就动机㊁工作不安全感的中介作用和上级发展性反馈的调节作用㊂通过对417份企业微观调研数据进行多元线性回归分析与Bootstrap 检验发现:第一,数字化转型可以激活工作成就动机,从而提升员工敬业度;第二,数字化转型亦会诱发工作不安全感,从而降低员工敬业度;第三,上级发展性反馈强化了数字化转型与工作成就动机之间的关系,弱化了数字化转型与工作不安全感之间的关系;第四,上级发展性反馈强化了数字化转型通过工作成就动机影响员工敬业度的间接路径,弱化了数字化转型通过工作不安全感影响员工敬业度的间接路径㊂研究结论揭示了数字化转型对员工情感的双刃剑效应,有助于企业员工的敬业度管理,并为数字化转型战略提供了有益的启示㊂关键词:数字化转型;工作成就动机;工作不安全感;上级发展性反馈;员工敬业度中图分类号:F270㊀㊀文献标志码:A㊀㊀文章编号:10002154(2024)02000512DOI:10.14134/33-1336/f.2024.02.001Boosting Morale or Creating Anxiety ?The Double-Edged Sword Effect of Digital Transformation on Employee s EngagementYI Lingfeng,YU Zhiyuan,CHEN Haoyu(School of Economics and Management ,East China Normal University ,Shanghai 200062,China )Abstract ︰Steaming from the conservation of resource theory,this article constructs a impact model of digital transformation on employee s engagement,analyzes the mediating role of job achievement motivation and job insecurity,and the moderating role of supervisor s developmental feedback.Through multiple linear regression and bootstrapping on 417micro survey data,the results show that:firstly,digital transformation enhances job achievement motivation,thereby improving employee s engagement;secondly,digitaltransformation reduces job insecurity,thereby reducing employee s engagement;thirdly,supervisor s developmental feedback moder-ates the relationship between digital transformation and job achievement motivation positively,while moderating the relationship be-tween digital transformation and job insecurity negatively;fourthly,supervisor s developmental feedback reinforces the mediating effect of digital transformation on employee s engagement through job achievement motivation,and weakens the mediating effect of digital transformation on employee s engagement through job insecurity.The findings reveal the double-edged sword effect of digital transfor-6商㊀业㊀经㊀济㊀与㊀管㊀理2024年mation on employees emotions,which contributes to the management of employees engagement in organizations and provides useful insights into digital transformation strategies.Key words︰digital transformation;job achievement motivation;job insecurity;supervisor s developmental feedback;employee s engagement一、引㊀言数字经济与传统行业的深度融合促使企业生态环境发生改变[1]㊂企业为了更好适应新环境的变化,逐步推进数字化转型战略,并将其视作组织创新升级和变革发展的枢纽㊂由此,引发了学界对于数字化转型相关问题的广泛探讨㊂研究表明,数字化转型在推动商业模式创新[2]㊁增加企业创新绩效[3]㊁提升全要素生产率[4]等方面均具有积极影响㊂文献梳理发现,数字化转型在人力资源管理领域的研究较少,数字化转型如何影响员工情感㊁态度和行为的机理也不明晰[5]㊂厘清数字化转型对员工敬业度的影响作用,有助于企业采取有效措施管理员工的情感,进而提升企业的绩效水平㊂本文讨论数字化转型对员工敬业度影响的双刃剑效应,即企业数字化转型在给员工带来积极体验的同时也可能对员工敬业度产生消极的影响作用㊂一方面,企业通过数字化转型向大数据㊁云计算及社交媒体平台进行组织结构转换[6],这不仅有助于激发员工的创新思维㊁实现更多价值创造㊁获取丰富的信息资源[7],还帮助员工从重复和烦琐的工作中解脱出来㊂员工在数字化转型的过程中体验到了更高的工作价值感和成就感[8],内在工作动机被激发的同时,提升了员工的敬业度㊂但另一方面,数字化转型也会改变企业原有运营模式和业务流程[9],导致员工所具备的传统知识技术与新颖的数字化业务形式之间的适配度下降,从而引发换岗,甚至失业的风险[7]㊂工作特征的变化增加了个体的职业危机,由此产生的职业未知性,可能会使员工变得烦躁㊁不安,以致无法专注于工作,从而降低员工的敬业度㊂本文认为,仅从单一视角关注数字化转型对员工敬业度的影响是不全面的,数字化转型在提振员工士气的同时,还可能会引发员工的心理压力㊁不适感等焦虑体验㊂因此,本文将讨论企业的数字化转型战略可能引发的双刃剑效应,探究数字化转型对员工工作态度和情感可能产生的差异化影响㊂资源保存理论认为,个体倾向于维系现有资源,并试图创造和获取新的资源[10],资源在流动中存在增益螺旋和损耗螺旋两条路径[11]㊂本文基于这两条路径构建数字化转型影响员工敬业度的双刃剑模型,讨论数字化转型对员工敬业度的差异化影响㊂一方面,工作成就动机作为有益的情感性资源,表征了增益螺旋,即数字化转型推动了线上业务的协同发展㊁打破了部门间信息系统的隔离㊁提升了资源的利用率,因此,员工工作效能和职业发展期望得以提升[12]㊂在工作成就动机的驱动下,员工以获取更高的绩效水平㊁实现个人价值为目标,增加了工作中时间与精力的投入,从而提升了敬业度㊂另一方面,工作不安全感作为认知资源消耗的衍生品,表征了损耗螺旋,即数字化转型加速了企业技术变革进程,部分工作岗位被智能技术替代的同时,对员工岗位技能也提出了更高要求[13]㊂工作的可替代性进一步催生了个体的工作不安全感,员工在心理资源遭受损耗的过程中,降低了工作热情,减少了工作时间与精力的投入㊂综上,基于资源保存理论,本文试图探究工作成就动机和工作不安全感在数字化转型影响员工敬业度之间的中介作用,阐明数字化转型对员工工作情感和态度的影响作用㊂此外,为揭示员工面对数字化转型产生不同感知的原因,本文基于外界资源获取的视角分析数字化转型影响员工敬业度的边界条件㊂资源保存理论提出,具备较多资源的个体更有能力获取资源,从而增加自身资源存量[14]㊂上级发展性反馈是激励且纠正员工行为的有效策略,其被定义为直属领导向下级提供有价值的信息,促使个体在工作中获得发展和进步的反馈方式[15]㊂有研究指出,组织领导为获取发展红利,会在数字化转型过程中,提供较多的上级发展性反馈[7]㊂然而,如果将上级发展性反馈视为外部资源,它是否会通过强化增益螺旋,增加数字化转型对工作成就动机的正向影响作用,抑或是弱化损耗螺旋,缓解数字化转型对工作不安全感的诱发作用?这是本文试图厘清的另一个重要问题㊂综上所述,本文基于资源保存理论,引入工作成就动机和工作不安全感作为中介变量,探析数字化转型影响员工敬业度的双刃剑效应,并将上级发展性反馈作为调节变量,分析数字化转型影响员工敬业度的边界条件㊂在企业数字化转型的背景下,本文的结论将为企业提升员工敬业度提供实践启示,丰富人力资源视角下企业数字化转型的本土研究㊂二㊁理论基础与研究假设(一)资源保存理论下数字化转型对员工敬业度的双刃剑效应数字化转型是指企业借助数字技术改变业务流程㊁内部文化和组织结构,提高自身运行效率进而满足不断变化的市场需求的商业模式[16]㊂员工敬业度是指个体在身体㊁认知及情感方面完成本职工作,并将展现自我与工作角色相结合的程度[17]㊂本文从组织整体变革视角出发,关注作为新兴变革方式的数字化转型对员工敬业度的影响效应㊂数字化转型不仅提升了员工的工作效率,同时也对个体数字化工作技能提出了更高要求,由此对员工敬业度产生的负面影响不容忽视㊂根据资源保存理论,资源是指个体能够感知到的㊁可以利用的有助于其实现工作目标的要素[18]㊂个体努力维系并培育有价值的资源,由于资源储备存在差异化,会形成增益螺旋和损耗螺旋两种影响路径㊂本文将以此为基础来说明数字化转型对员工敬业度的双刃剑效应㊂就增益螺旋路径而言,数字化转型拓宽了员工数据信息来源,不仅有助于个体实现更多价值创造,而且使个体从重复㊁规范的程序化工作中解脱出来,提升了工作效率,促使其对未来职业发展充满信心,激发自我提升动机,进而愿意在工作中投入更多,提升了敬业度㊂就损耗螺旋路径而言,数字化转型重组了原有工作岗位的任务内容,对员工提出了更高的角色期待和技能要求,并且导致部分工作岗位被智能化机器和系统所替代[19],个人岗位层级与存续性受到威胁,产生的焦虑㊁不安等情绪导致员工不愿在工作中投入更多时间与精力,可能对员工敬业度产生负面影响㊂(二)数字化转型与员工敬业度:通过工作成就动机的增益螺旋机制工作成就动机作为重要的社会性动机,被定义为个体保持兴趣㊁愉悦和高度自信的状态来克服困难或进入具有挑战性工作情境的倾向,旨在满足个人既定需求[20]㊂根据资源保存理论中的增益螺旋机制,数字化转型构筑的智能办公模式提升了工作效率,促使个体对工作充满信心,心理资源更加充裕,有助于工作成就动机的建立,这种积极的情感资源激励员工以更加饱满㊁热情的态度投入工作,从而提升员工的敬业度㊂首先,关注数字化转型对工作成就动机的影响㊂通过数字化转型能够变革企业运营模式和组织形式,加速了多维数据信息向组织的渗透,提升了信息传播的广泛性和效率[21-22]㊂部门之间信息交流变得愈加高效和频繁,这不仅有助于员工依靠信息资源优势完成更多价值创造,而且数字化工作场景还帮助员工从重复且烦琐的工作中解脱出来㊂个体能够自主安排工作流程,促使其体验到更高的工作价值感[8],在提升工作期望,增加心理资源的同时,产生了强烈的工作成就动机㊂基于此,本文提出:假设1:数字化转型正向影响工作成就动机㊂其次,关注工作成就动机对员工敬业度的影响㊂资源保存理论提出,拥有较多资源的个体通常更具工作活力[23]㊂工作成就动机作为有益的情感性资源,一定程度上激发了个体活力,促使其产生强烈的成功需求,并倾向于接受更有挑战性的任务[24]㊂Lepine 等[25]认为,在面对既定目标所带来的挑战性压力时,高工作成就动机者相较低工作成就动机者会付出更多的时间和精力来促进自身发展,并期待任务完成后获得成就感㊂因此,他们会专注于学习有关任务完成的新知识与新技能,追寻卓越和完美,以积极态度投入工作,从而获取丰富的工作资源,更好保障目标的达成,进一步提升了员工敬业度㊂基于此,本文提出:假设2:工作成就动机正向影响员工敬业度㊂综上,数字化转型使企业内部的信息交流变得更加高效和广泛,不仅减少了员工之间的非必要接触,而且有助于员工进行更多价值创造,提升了个体工作效率,从而激发员工在职业生涯中的工作成就动机㊂7㊀第2期㊀易凌峰,余志远,陈浩宇:提振士气还是制造焦虑?数字化转型对员工敬业度的双刃剑效应8商㊀业㊀经㊀济㊀与㊀管㊀理2024年工作成就动机作为一种相对稳定的内在驱动力,一定程度上促使员工积极学习与目标相关的新知识与新技能,为了进一步保障目标的达成,员工会在工作中投入更多的时间和精力,这进一步提升了员工敬业度㊂基于此,本文提出:假设3:数字化转型通过工作成就动机正向影响员工敬业度㊂(三)数字化转型与员工敬业度:通过工作不安全感的损耗螺旋机制工作不安全感作为个体主观感知,其被定义为员工体验到自身工作或重要工作特征受到威胁所萌发的担忧感[26]㊂根据资源保存理论中的损耗螺旋机制,数字化转型的推进,不仅对员工的数字工作技能提出了更高要求,需要其在认知及情感方面持续投入更多精力,造成了个体心理资源的损耗,而且数字化转型构筑的智能办公模式还替代了部分工作岗位,导致员工面临失业危机,由此引发的工作不安全感挫伤了个体积极性,使其无法专注于工作,从而造成员工敬业度的降低㊂首先,关注数字化转型对工作不安全感的影响㊂数字化转型改变了组织结构及战略[27],使得个体工作形态被重新定义,这不仅对员工提出了更高的岗位技能要求[28],迫使员工更加关注自身的技能发展[29],而且随着转型的不断推进,对专业技能门槛和岗位层级形成了潜在壁垒㊂低端劳动力将会被数字技术所代替,员工面对失业风险会进一步产生焦虑㊁不安等负面情绪,甚至发生严重的情绪内耗,致使个体工作热情和工作投入进一步减少[30],工作不安全感由此变得更加强烈㊂基于此,本文提出:假设4:数字化转型正向影响工作不安全感㊂其次,关注工作不安全感对员工敬业度的影响㊂敬业度是指个体对工作热情㊁认真以及投入的程度[31]㊂由此可见,敬业度与个体所投入智力㊁精力以及努力密切相关㊂根据资源保存理论,工作不安全感的产生削弱了个体就业身份认同,致使其在寻找和获取信息层面耗费了大量资源和精力,从而没有更多资源和精力投入本职工作[23],对敬业度产生了不利影响㊂此外,工作不安全感还会诱发员工惶恐㊁不安等消极情绪,损耗了个体心理资源,使员工不能专注于自身工作[32],从而导致了员工敬业度的降低㊂基于此,本文提出:假设5:工作不安全感负向影响员工敬业度㊂综上,数字化转型使得个体工作形态被重新定义,对员工的数字化工作技能提出更高要求,部分低端劳动力将逐渐被数字技术所代替,从而增加了员工的失业风险,个体工作不安全感会由此变得更加强烈㊂工作不安全感的产生不仅挫伤了个体积极性,使其面对失业风险产生焦虑㊁不安等负面情绪,还损耗了个体心理资源,使其无法将全部精力投入本职工作,从而降低了员工敬业度㊂基于此,本文提出:假设6:数字化转型通过工作不安全感负向影响员工敬业度㊂(四)上级发展性反馈的调节作用资源保存理论提出,个体所具备的资源水平对其在特定情境下的心理资源变化过程存在一定影响[23]㊂当个体拥有较为丰富的资源时,不仅更容易获取新的资源,而且对已经损耗的资源具有弥补作用[33]㊂因此,本文认为上级发展性反馈作为重要的组织资源,会通过改善个体资源拥有状况,激发员工工作的主动性和建设性[34],促使其以更加积极的态度应对数字化转型,从而增强数字化转型引发的增益螺旋机制,削弱数字化转型引发的损耗螺旋机制㊂根据资源保存理论,本文认为上级发展性反馈有助于强化数字化转型对工作成就动机的影响㊂具体而言,当上级发展性反馈水平较高时,经历数字化转型的员工能够及时接收到其工作所需的信息资源,使他们对数字化工作目标及内容具有清晰的认知,也能够更有针对性地改进不足,提升数字化工作技能,增加自身心理资源,形成资源增益螺旋,从而以更加积极的态度应对数字化转型,逐步感受到数字化工作模式所带来的便捷性和长期效益,由此产生对工作的积极期望㊂在这种积极期望的激励下,个体会产生更多正向的工作动机[35],并对自身职业发展充满信心,工作成就动机被更为强化地激发㊂反之,如果上级发展性反馈水平较低则会导致员工在数字化转型过程中遇到的疑惑和困难无法得到及时回应,从而对个体工作的有效开展造成阻碍,致使其产生紧张㊁不安的情绪,降低对自身职业前景的期待,这不利于工作成就动机的建立㊂基于此,本文提出:假设7:上级发展性反馈正向调节数字化转型与工作成就动机间的关系㊂根据资源保存理论,本文认为上级发展性反馈有助于弱化数字化转型对工作不安全感的影响㊂具体而言,上级发展性反馈本质上是信息反馈,它可以为员工提供有用的信息,而不是做出强制性的工作要求来提升他们的绩效[15]㊂当上级发展性反馈水平较高时,经历数字化转型的员工能够及时接收到有助于其提升数字工作技能的建设性信息,与此同时,高水平的上级发展性反馈能够显著增强员工的自我效能感,促使员工改善工作态度和行为[36],并尽快适应新的数字化工作模式,个体心理资源也进一步得到丰富,降低了数字化转型引发工作不安全感的可能性㊂反之,如果上级发展性反馈水平较低,忽略或延迟回应员工在数字化转型过程中遇到的疑惑,将进一步诱发个体工作角色的模糊性,这不仅不利于员工开展工作,而且随着时间压力的增加,员工紧张㊁不安的情绪会加剧[37],从而萌发更加强烈的工作不安全感㊂基于此,本文提出:假设8:上级发展性反馈负向调节数字化转型与工作不安全感间的关系㊂(五)被调节的中介作用综合假设3与假设7,当上级发展性反馈水平较高时,经历数字化转型的员工能够及时获取上级反馈和工作资源,工作成就动机得以更为强化地激发,从而愿意在数字化转型的工作模式中持续投入更多精力,更加爱岗敬业,提升了员工敬业度;综合假设6和假设8,当上级发展性反馈水平较高时,经历数字化转型的员工能够及时获取上级支持进而减少了工作不安全感的产生,使得数字化转型通过工作不安全感负向影响员工敬业度的路径被削弱㊂基于此,本文提出:假设9:上级发展性反馈正向调节数字化转型通过工作成就动机影响员工敬业度的间接路径㊂假设10:上级发展性反馈负向调节数字化转型通过工作不安全感影响员工敬业度的间接路径㊂综上所述,本文构建如图1所示的理论模型:图1㊀理论模型三㊁研究设计(一)样本选取与数据收集本文通过向甘肃㊁陕西㊁四川等三个地区的40家高科技企业员工发放线上问卷,以获取样本数据㊂所调研的企业主要涉及通信技术和智能制造等新兴产业,选择这些产业的原因在于上述行业企业的数字化转型进程较快,同时,调研企业均具有数字化转型的背景和数字化转型的工作任务,样本具有代表性㊂为了降低共同方法偏差的影响,数据收集工作分为3个时点,每次数据收集的间隔时间为一个月左右,要求被试填写手机号后四位,同时借助作答软件记录被试的微信名及用户名,从而保证3个时点的测量数据能够进行有效匹配㊂T1(时间点1)阶段收集了控制变量㊁数字化转型以及上级发展性反馈样本数据;T2(时间点2)阶段收集了工作成就动机和工作不安全感样本数据;T3(时间点3)阶段收集了员工敬业度样本数据㊂在T1阶段,共计回收问卷553份,剔除51份无效问卷后获得502份有效问卷㊂在T2阶段,将问卷精确推送到T1阶段形成的样本库中,通过手机号后四位㊁微信名及用户名匹配,共计回收447份有效问卷㊂在T3阶段,通过精确推送和仔细筛选最终形成417份有效问卷,问卷有效回收率为75.4%㊂性别方面,男性占比9㊀第2期㊀易凌峰,余志远,陈浩宇:提振士气还是制造焦虑?数字化转型对员工敬业度的双刃剑效应48.7%,女性占比51.3%;年龄方面,25岁及以下占比21.1%,26 35岁占比19.9%,36 45岁占比19.4%, 46 55岁占比20.4%,55岁以上占比19.2%;学历方面,本科以下占比40.5%,本科占比39.1%,硕士占比15.1%,博士占比5.3%;婚姻状况方面,已婚占比40.0%,未婚或离异占比60.0%;工作年限方面,不足1年占比26.1%,1 3年占比30.0%,4 6年占比20.6%,7 9年占比22.3%,9年以上占比1.0%;企业性质方面,国有企业占比28.1%,民营企业占比23.9%,中外合资企业占比24.7%,外资企业占比23.3%㊂(二)测量工具为进一步提升测量工具的可靠性,本文选取国内外较为经典权威的量表,采用李克特5点计分法进行测量,其中, 1=非常不符合;2=不符合;3=一般符合;4=比较符合;5=非常符合 ㊂1.数字化转型㊂数字化转型采用了胡青[38]所研发的量表,共计5个测量条目,例如 本企业采用数字技术对现有产品㊁服务和流程进行改造升级 等㊂2.工作成就动机㊂工作成就动机采用了Steers和Braunstein[39]所研发量表的成就需要部分,共计5个测量条目,例如 当工作任务相当艰巨的时候,我也会竭尽全力 等㊂3.工作不安全感㊂工作不安全感采用了Hellgren等[40]所研发的量表,共计7个测量条目,例如 在未来的一段时间里,我可能有不得不离开现在工作的风险 等㊂4.上级发展性反馈㊂上级发展性反馈采用了Zhou[15]所研发的量表,共计3个测量条目,例如 我的上司会给我提供有助于我发展的信息 等㊂5.员工敬业度㊂员工敬业度采用了张轶文和甘怡群[41]基于中国情境修订后的UWES-9量表,共计9个测量条目,例如 我对自己的工作充满了热情 等㊂6.控制变量㊂本文将性别㊁年龄㊁学历㊁婚姻状况㊁工作年限和企业性质设为控制变量,提升研究结论的可信度㊂四㊁实证分析(一)共同方法偏差检验本文采用Harman单因素检验方法,将数字化转型㊁工作成就动机㊁工作不安全感㊁上级发展性反馈以及员工敬业度的测量条目置于一起,进行未旋转的因子分析㊂首个因子解释变异量为29.133%,远小于40%的标准要求,共同方法偏差在合理范围内㊂同时,为了进一步检验共同方法偏差问题是否会对接下来的数据分析产生影响,我们也采用了潜在误差变量控制法对样本数据进行了检验㊂结果显示,在加入共同方法因子后,拟合指数的变化值为:ΔRMSEA=0.002,ΔCFI=0.002,ΔTLI=0.002,ΔSRMR=0.006,均小于0.05的标准要求,因此,共同方法偏差问题通过检验㊂(二)信度与效度分析本文分析变量信度㊁聚合效度以及区分效度㊂由表1可见,所有测量因子的标准化因子载荷均大于0.7,数字化转型㊁工作成就动机㊁工作不安全感㊁上级发展性反馈以及员工敬业度的Cronbach sα值和CR值大于0.8,AVE值大于0.5,表明信度和聚合效度合格㊂表1㊀变量信度与聚合效度检验结果变量测量因子标准化因子载荷Cronbach sαCR AVE数字化转型DT10.760DT20.794DT30.741DT40.821DT50.8320.8910.8930.62501商㊀业㊀经㊀济㊀与㊀管㊀理2024年。