Resources Allocation Among Heterogeneous DMUs

With the quick economic integration paces,the competitions among the companies become more and more furious.It is well known that resource is the material basis for company's survival and development.While the resource-deploying ability is the core competitiveness of companies and the fundamental guarantee for their sustainable development.Along with technology advancement and production socialization,the scale of company has been continually expanding and their business increasingly diversifying.Hence,faced with such complicated organization structure,how to allocate resource efficiently and reasonably has become a critical problem to be settled for many companies.Also,the issue of resource allocation is a dominant theme in management science,which roused considerable attentions and efforts from more and more scholars.Recently,the wide use of data envelopment analysis(DEA)brings a new flavor to the problem,because it provides decision makers with an effective approach to devise mechanisms that would improve resource allocation from the perspective of operational efficiency.Based on the research outcome so far,this paper focuses on company-wide resource allocation with considering the presence of heterogeneity in decision-making units(DMU).The study results can provide effective decision supports and significant reference for company managers.This paper mainly divided into five chapters,and the corresponding contents are concluded as follows:The first chapter is an introduction where the background,the research problem,the basic theory of DEA,the status,the contents and the academic value are presented.The second chapter introduces a new DEA approach to allocate the resource in the branch network system which is typically appears in multi-national or multi-regional companies with many branches in different locations.Considering the technology heterogeneity that caused by the difference in branch's locations,we introduces group-specific technologies to allocate resource for such company with assumption that DMUs in the same regions have similar technology.Given the spatial distribution of the production,we impose additional restrictions on resource flows and divide the resource inputs into three groups:fixed inputs,regional inputs that allocated to the DMUs in the same area and common resource that allocated freely to all DMU.Furthermore,the objective of the proposed models is to maximize the gross profits of the company which is a natural assumption for the for-profit organization.At last,an example was presented to illustrate the proposed approach with heterogeneous technology is more practically feasible and superior than the prior approach with homogeneous technology.On the basis of chapter 2,the third chapter studies how to allocate resource and assort the regional production to meet market demand as much as possible.The previous researches related to company-wide resource allocation mainly focused on how to allocate resource to maximize the total production from the efficiency perspective,which departs from market mechanism's regulation.They neglected that excess capacity or insufficient capacity has negative impact on the benefit of companies in the markets.In addition,the production capability of branches in different regions may not match with the located demand.Hence,based on the demand fluctuations in different regions,we develop a new resource allocation DEA model that allows the outputs of a region swift to other regions to realize the effective production of the entire company,i.e.to balance the production of company and the total market demand by strengthening the regional cooperation.Chapter 4 extends the traditional resource allocation problem to situations where the DMUs have multiple heterogeneous processes based on network DEA method.Different with traditional DEA views DMUs as "black-box",we investigate the internal structure of DMUs,and divide the production system of DMUs into multiple heterogeneous sub-units.Then we define the production possibility set and transformation possibility set for each group of sub-units.We develop different DEA models for different situations as the sub-units are operated independently,sub-units have shared inputs,the sub-units are linked each other with intermediates and some intermediates exit to allocate resource to maximize the total outputs.Our models can provide the detail information of configuration processes between the DMUs and the sub-units.And according to the optimal distribution plan can help the managers to set output targets for each sub-unit to enhance the management.The last chapter concludes all the work and innovations of this dissertation.Based on the analysis on the deficiency of this dissertation,the future directions are also provided.The main contributions of this paper include as:Firstly,this paper extends studies the problem of resource allocation in the multi-national or multi-regional company with branch network system,due to the difference in the geographical locations we generalize the model to accommodate technological heterogeneity and restrict the resource flows between different regions.Secondly,we present new approaches to allocate resource and assort the regional production to promote the balance of supply and demand based on effectiveness measures.Finally,this paper studies how to allocate resource between DMUs and sub-units based on network DEA method,which extends the traditional problem of resource allocation to situations where the DMUs have multiple heterogeneous processes.