Abstract:Compared with traditional auctions , online auctions have some unique characteristics , which in clude bidders ’ stochastic arrival , snipping effects , the existence of auction platform and the fee charged to sellers , as well as multi unit and multiple stages. Taking into account these unique characteristics , this paper explores some important questions of multi unit online auctions systematically. The result shows that goods with larger valuation disparity are more suitable to be on sale by auction but under the precondition that the network user basement should achieve a certain scale ;under a unified analysis framework , we solve the seller ’ s opti mal public reserve price and private reserve price respectively , and also verify that the seller with private re serve price can gain more profits than the seller with public reserve price ;considering the inconsistency be tween the auction platform and the seller ’ s objectives , we indicate that , to decrease the listing fee , increasing ommission ratio and shortening the auction duration are both helpful for us to design the incentive compatible optimal auction mechanism ;to auction more goods is not always optimal for the seller , we solve for the optimal number of goods to be sold in one auction under one stage and multiple stage auctions respectively.

Abstract:Knowledge diffusion can be accelerated through the network. This paper constructs a knowledge diffusion model of high tech enterprise alliance based on NW small world network to reveal the characteristics of the alliance ’ s knowledge diffusion , and uses MATLAB software for simulation. The study indicates that re duction of characteristic path length of network , increase of network cluster coefficient and improvement of knowledge communication frequency between the members are active measures for promoting knowledge diffu sion of high tech enterprise alliances , enhancing innovation effectiveness and developing knowledge innovation.

Abstract:The problem of scheduling parallel batch processing machines is considered from a clustering per spective in this paper. We first demonstrate that the batching problem with non identical job sizes can be re garded as a generalized clustering problem , providing a novel insight into scheduling with batching. The con cept of WR (waste ratio of batch space )is then presented and the objective function of minimizing makespan is transformed into minimizing weighted WR so as to define the distance measure between batches in a more understandable way. The equivalence of the two objective functions is also proved. In addition , a clustering algorithm CACB (constrained agglomerative clustering of batches )is proposed based on the definition of WR to generate batches. The experimental results show that CACB outperforms the existing approaches BFLPT (best Fit longest processing time )and GA (genetic algorithm )in large scale problems.

Abstract::Supply chain logistics planning is the key issue of supply chain management. To solve the logistics planning problem of decentralized supply chain , in which the nodes can be unlimitedly expanded , the concept of supply chain cell is put forward. Considering the randomicity of supply and demand price among the supply chain , a integrated logistics planning model for the multi stages , multi nodes , multi products and decentral ized supply chain is introduced based on the stochastic chance constrained programming theory. Based on ran dom simulation , a hybrid intelligence algorithm is developed for this model. With a numerical examples simu lation the effectiveness of the model and the algorithm is proved , meanwhile the reasonable ranges of parameter values in the algorithm are defined and the sensitivity of the parameters in the model are analyzed with numerical example simulation.

Abstract:A supply chain consisting of one manufacturer , one seller , and a group of customers with strategic behavior and risk preference is considered in this paper. Under strategic customer behavior and risk prefer ence , the seller ’ s pricing and inventory decisions are analyzed. Supply chain profit sharing contract is studied under rational expectations equilibrium. Results show that the seller ’ profit decreases as customer ’ s risk pref erence increases , and the seller would have to lower product price and initial inventory so as to mitigate profit reduction. The wholesale price of the profit sharing contract should be set above manufacturer ’ s marginal pro duction cost because of the existence of strategic customers. Moreover , the sharing ratio parameter within the profit sharing contract benefits redistribution of the total profit of the supply chain. The bargaining powers of supply chain members determine the sharing ratio parameter.

Abstract:There has been abundant research on service failure and service recovery. Surprisingly , most of the papers are from an individual consumer perspective , and not a group consumer perspective. However , group consumer failures occur rapidly year by year in China. Compared with individual failure , interaction of group consumers is more complicated and more variable , so that it is more difficult for Service Corporation to cope with. By the use of real experiment , the authors discuss the impact of deindividuation and group emotional contagion on customer ’ s negative emotion under group service failure setting. Results indicate that both group emotional contagion and deindividuation effect both exists in group consumers and which will bring direct impact on customer ’ s complaint intention.

Abstract:By analyzing the C2C sales records , we find the repeat purchase frequency of customers is power law distribution , which is different from the result of existing repeat purchase research that the purchase fre quency is negative binomial distribution. Based on actual situation in C2C transactions , a repeat purchase model with buying reinforcement is built , which explaned to the phenomenon of the power law distribution. Our model and empirical analysis show that , in C2C environment , the distribution of customers ’ repeat pur chase frequency is composed of a long tail and short head , meaning that a large number of customers have a low number of purchases , while a few customers have a high repeat purchase frequency. Our model also re veals that the huge number of potential customers and the repeat purchase with buying reinforcement contrib utes to the power law distribution characteristics of repeat purchase frequency.

Abstract:In this paper , based on the 2SLS residuals in the spatial econometric autoregressive model , we prove that Moran test is asymptotically normal distribution when the error is independent and identically dis tributed , and then establish OLLMoran test. Monte Carlo experiment results show that size distortion of OLL Moran test in this research is less than that of KP Moran , and the power of OLL Moran test is more than that of KPMoran. OLL Moran test has good finite sample performance , and could check effectively spatial correlation among 2SLS residuals in the spatial econometric autoregressive model.

Abstract::This paper studies the effect of property tax on the residential real estate market , considering the strategic behavior of consumers , investors and real estate developers. Based on a simulation model , we analyze the different effects of property tax and exercise taxes on the behavior of consumers and real estate developers. We obtain the market equilibrium price and quantity in our model. This paper shows that property tax plays a more important role than any other taxes on housing price control. In addition , exercise tax on sale plays a bet ter role than exercise tax on rent. However , exercise taxes can only be considered as a temporary policy for housing price control. For the long run social welfare , given the underlying market condition , policy maker should consider property tax with necessary supporting measures in order to meet the corresponding basic resi dential demand.