ABSTRACT A DEA-BASED APPROACH TO RANKING MULTI-CRITERIA ALTERNATIVES Tuncer, Ceren M.Sc., Department of Industrial Engineering Supervisor: Prof. Dr. Murat K&ouml / ksalan August 2006, 88 pages This thesis addresses the problem of ranking multi-criteria alternatives. A Data Envelopment Analysis (DEA)-based approach, the Method of the Area of the Efficiency Score Graph (AES) is proposed. Rather than assessing the alternatives with respect to the fixed original alternative set as done in the existing DEA-based ranking methods, AES considers the change in the efficiency scores of the alternatives while reducing the size of the alternative set. Producing a final score for each alternative that accounts for the progress of its efficiency score, AES favors alternatives that manage to improve quickly and maintain high levels of efficiency. The preferences of the Decision Maker (DM) are incorporated into the analysis in the form of weight restrictions. The utilization of the AES scores of the alternatives in an incremental clustering algorithm is also proposed. The AES Method is applied to rank MBA programs worldwide, sorting of the programs is also performed using their AES scores. Results are compared to another DEA-based ranking method. Keywords: Ranking, data envelopment analysis, weight restrictions.

Today&rsquo / s competitive environments and heightened expectation of customers make it necessary to improve quality of products and processes continuously. Therefore, quality improvement is a major concern for companies. Determining improvement priorities for not only long but also short term bottom line results is a key problem in quality improvement management. In this thesis a practical decision matrix based method is developed for selecting quality improvement projects by considering throughput and quality loss in manufacturing environments with inspection error and rework. Performance of the proposed method under different experimental conditions is analyzed and results are discussed.

This study aims to improve component placement sequencing of a number of PCBs produced on a turret style SMT machine. After modeling the problem and having found that an optimal solution to the real PCB problem is hard to be achieved because of the concurrent behavior of the machine and the PCB design parameters, two heuristics are developed by oversimplifying the problem down to TSP. Performance of the heuristics and the lower bounds is evaluated by comparing the results with the optimal solution for two sets of randomly generated PCBs. The heuristic solutions are also compared with the lower bounds and the current implementation for the real PCBs. It is found out that the heuristics improve the current efficiency figures of the company.

In this thesis, we deal with the demand driven disassembly planning. The main aim of the study is to construct heuristic approaches according to the suggested improvements in the literature. These heuristic approaches are further improved by recognizing the key points of the disassembly planning problem. All of the solution approaches aim minimizing total cost related to relevant costs of disassembly operations. Another subject given attention in this thesis is the importance of the setup cost on the disassembly planning, which has not been studied yet in the literature to the best of our knowledge. Computational studies are carried out to assess the performance of the heuristic procedures proposed.

In this study, the seat inventory control problem is considered for airline networks from the perspective of a risk-averse decision maker. In the revenue management literature, it is generally assumed that the decision makers are risk-neutral. Therefore, the expected revenue is maximized without taking the variability or any other risk factor into account. On the other hand, risk-sensitive approach provides us with more information about the behavior of the revenue. The risk measure we consider in this study is the probability that revenue is less than a predetermined threshold level. In the risk-neutral cases, while the expected revenue is maximized, the probability of revenue being less than such a predetermined level might be high. We propose three mathematical models to incorporate the risk measure under consideration. The optimal allocations obtained by these models are numerically evaluated in simulation studies for example problems. Expected revenue, coefficient of variation, load factor and probability of the poor performance are the performance measures in the simulation studies. According to the results of these simulations, it shown that the proposed models can decrease the variability of the revenue considerably. In other words, the probability of revenue being less than the threshold level is decreased. Moreover, expected revenue can be increased in some scenarios by using the proposed models. The approach considered in this thesis is especially proposed for small scale airlines because risk of obtaining revenue less than the threshold level is more for this type of airlines as compared to large scale airlines.

Vendor Managed Inventory (VMI) is a business practice in which vendors monitor their customers&rsquo / inventories, and decide when and how much inventory should be replenished. VMI has attracted a lot of attention due to its benefits. In this study, we analyze the benefits of VMI in a supply chain consisting of a single retailer and a single capacitated supplier under stochastic demand. We propose a VMI setting and compare the vendor managed system with the traditional system to quantify the benefits of VMI. In our proposed VMI system, the retailer shares the inventory level information with the supplier, which is not available in traditional system / and the supplier is responsible to keep the retailer&rsquo / s inventory level between the specified minimum and maximum values, called (z,Z) levels, set by a contract. We examine the benefits of such a VMI system for each member and for the overall chain / and analyze the effects of system parameters on these benefits. The performance of VMI in coordinating the overall chain is examined under different system parameters.

In this study, multicriteria sorting methods based on Data Envelopment Analysis (DEA) are developed to evaluate industrial Research and Development (R&amp / D) projects proposed to Technology and Innovation Grant Programmes Directorate (TEYDEB) of the Scientific and Technological Research Council of Turkey (T&Uuml / BITAK). Even though DEA is used extensively as a multicriteria decision making (MCDM) tool for ranking / to our knowledge, this study is the first attempt utilizing DEA for sorting purpose. A five level R&amp / D project selection criteria hierarchy and an assisting point allocation guide with a scale of ten-points are derived to measure and quantify the performance of the proposals. The interval pairwise comparison matrices determined from the judgments of TEYDEB managers are used to obtain weight intervals from Analytic Hierarchy Process (AHP) model. These weights are employed as assurance region constraints. Motivated from the fact that derived criteria constitute inputs and outputs of R&amp / D projects / DEA determining efficiencies based on inputs and outputs is utilized for sorting. Based on this approach, two threshold estimation models, PM1 and PM2, and five assignment models, APM1, APM2, APM3, APM4 and APM5, are proposed. The models are applied to a case study in which 60 projects are placed into four groups according to two reference sets composed of proposals from the year 2009. The well-known muticriteria sorting method, UTADIS, is also implemented for comparison. It is concluded that proposed methods are more stable than UTADIS and the integrated application of threshold estimation model PM2 and assignment model APM4 provides the best results.

In this thesis, we work on adisassembly lot-sizing problem for multiple products with parts commonality,i.e., general product structure. We assume that supply of discarded products is infinite. When a product (or a subassembly) is disassembled, all its immediate child items are obtained,i.e., complete disassembly case.Intermediate and leaf items obtained are demandedbyexternal suppliers or remanufacturers. The maximum possible salesfor each intermediate and leaf item are known.Sales of the intermediate and leaf items are the revenue sources. The discarded products are purchased ata unit purchasing cost. The disassembly operation incurs a fixed and a variable disassembly cost. Due to this cost structure, intermediate and leaf items can be stocked incurring an inventory holding cost. We develop an integer programming formulation to determine the time and quantity of the discarded products to be purchased / thetime and quantity of the discarded products and the intermediateitemsto be disassembled / and the time and quantity of intermediate and leaf items to be soldin order tomaximizethe total profit over a finite planning horizon. We state that ourproblem is NP-hard by refering the study of Kim et. al. (2009). We propose a heuristic solution approach that solves the problem in a reasonable computational time and generates near optimal solutions. The solution approach is based on the idea of sequentially solving a relaxed version of the problem and one-period integer programming models.In a computational study, the performance of the heuristic approach is assessed for a number ofrandomly generated problem instances.The results of the computational study show that the solutions of the heuristic approacharevery close to the optimal and the best feasible solutions obtained within the time limit.

In this study, we consider a resource investment problem with time/resource trade-offs in project environments. We assume each mode of an activity is characterized by its processing time and resource requirement and there is a single renewable resource. Our aim is to minimize the maximum resource usage, hence the total amount invested for the single resource. We formulate the problem as a mixed integer linear model and find optimal solutions for small sized problem instances. We propose several lower bounding procedures to find high quality estimates on the optimal resource investment cost. We use our lower bounds to evaluate the performance of our heuristic procedures. The results of our computational experiments have revealed the satisfactory performances of our lower bounds and heuristic procedures. Projects, Resource Investment Time/Resource Trade-off, Bounding Procedures

No description available.