Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
4 results
Search Results
Item Tracking the Cost Heuristic: A Rule of Thumb in Choice of Counterfinal Means(2013) Klein, Kristen; Kruglanski, Arie W; Psychology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Domain-specific empirical evidence shows that people sometimes infer a correlation between the cost of an item or behavior (e.g., price, effort) and its quality or efficacy (e.g., Labroo & Kim, 2009; Shiv, Carmon, & Ariely. 2005). In the present research, I proposed and tested a general mechanism that may underlie these instantiations. I termed this mechanism the cost heuristic, whereby people may infer that if a means is costly or detrimental to alternative goal(s) (i.e., counterfinal), then it must be highly instrumental to whatever focal goal it serves. Three preliminary studies provided inconsistent support for the cost heuristic. However, two of these studies suggested that a reverse halo effect (Thorndike, 1920) might have interfered with the cost heuristic. In three additional studies, I controlled for this reverse halo effect and tested more nuanced hypotheses about conditions under which the cost heuristic might be particularly likely to emerge. In Study 1, after statistically controlling for reverse halo effects, I found marginal support for a general cost heuristic, but not for the proposed moderators of alternative goal magnitude and the perceived ecological validity of the cost heuristic. In Study 2, the alternative goal magnitude manipulation was only effective in one goal domain; however, in this domain the results fully supported the hypothesis. Those who perceived the cost heuristic as ecologically valid were more likely to exhibit it under high (vs. low) alternative goal magnitude, whereas for those who did not perceive it as ecologically valid, alternative goal magnitude did not make a difference. In Study 3, the results did not support my hypothesis, but rather suggested that the vignettes' length/complexity may have obscured the detrimentality cue on which the cost heuristic is based. Taken together, the evidence from these studies is suggestive but inconclusive with regard to a cost heuristic and the conditions under which it might manifest. Limitations and future directions are discussed.Item THE VEHICLE ROUTING PROBLEM WITH DEMAND RANGES(2009) Cornick, Namrata Uppal; Golden, Bruce; Applied Mathematics and Scientific Computation; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The classic Capacitated Vehicle Routing Problem (CVRP) has been studied in the Operations Research field for over 5 decades. This thesis formulates the vehicle routing problem with a variation that has not been studied in detail. It is called the Vehicle Routing Problem with Demand Ranges (VRPDR). With increasing competition, corporations are looking to minimize costs. This problem aims to reduce the cost of distributing goods by allowing flexibility in the delivered or dropped off quantity. This benefits the customer as well, by reducing storage and other inventory costs. We solve the VRPDR problem where the customer gives the distributor a demand range. The distributor is rewarded for delivering more. A metaheuristic, record-to-record travel with demand range (RTRDR), is developed which is capable of solving large problem instances. The metaheuristic is a modification of a successful CVRP metaheuristic used in the past. In this thesis, we report results on problems ranging in size from 560 to 1200 customers. The developed metaheuristic uses the Clarke-Wright procedure to get initial solutions and then applies record-to-record travel in conjunction with two-opt moves, one point moves, and two point moves. Since the problem has not been studied yet from a computational point of view, we have developed a comparison algorithm, which takes advantage of the demand range flexibility of this problem only after the algorithm has optimized for distance alone. We use the results from this algorithm as a benchmark to compare with our proposed metaheuristic RTRDR.Item A Study of Four Network Problems in Transportation, Telecommunications, and Supply Chain Management(2007-08-01) Chen, Si; Golden, Bruce; Raghavan, Subramanian; Business and Management: Decision & Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The increasing material costs and the rapid advances in computing technology have both motivated and promoted the study of network problems that arise in several different application domains. This dissertation consists of four chapters on network applications in transportation, telecommunications, and supply chain management. The core of our research is to apply heuristic search procedures and combinatorial optimization techniques to various practical problems. In the second chapter we investigate the split delivery vehicle routing problem (SDVRP), where a customer's demand can be split among several vehicles. The third chapter deals with the regenerator location problem (RLP) that arises in optical networks. The fourth chapter solves the parametric uncapacitated network design problems on series-parallel graphs, which have potential application in supply chain management. In the fifth chapter we study the arc routing problem that arises in the small package delivery industry. The last chapter summarizes the dissertation. The results in this dissertation indicate that the methodologies developed to solve the network problems in the four different applications are quite efficient. Consequently, when applied in practice, they have the potential to significantly improve the operational efficiency of organizations in the relevant application domains.Item INTEGRATED PRODUCTION-DISTRIBUTION SCHEDULING IN SUPPLY CHAINS(2005-05-09) Pundoor, Guruprasad; Chen, Zhi-Long; Decision and Information Technologies; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)We consider scheduling issues in different configurations of supply chains. The primary focus is to integrate production and distribution activities in the supply chain in order to optimize the tradeoff between total cost and service performance. The cost may be based on actual expenses such as the expense incurred during the distribution phase, and service performance can be expressed in terms of time based performance measures such as completion times and tardiness. Our goal is to achieve the following objectives: (i) To propose various integrated production-distribution scheduling models that closely mirror practical supply chain operations in some environments. (ii) To develop computationally effective optimization based solution algorithms to solve these models. (iii) To provide managerial insights into the potential benefits of coordination between production and distribution operations in a supply chain. We analyze four different configurations of supply chains. In the first model, we consider a setup with multiple manufacturing plants owned by the same firm. The manufacturer receives a set of distinct orders from the retailers before a selling season, and needs to determine the order assignment, production schedule, and distribution schedule so as to optimize a certain performance measure of the supply chain. The second model deals with a supply chain consisting of one supplier and one or more customers, where the customers set due dates on the orders they place. The supplier has to come up with an integrated production-distribution schedule that optimizes the tradeoff between maximum tardiness and total distribution cost. In the third model, we study an integrated production and distribution scheduling model in a two-stage supply chain consisting of one or more suppliers, a warehouse, and a customer. The objective is to find jointly a cyclic production schedule at each supplier, a cyclic delivery schedule from each supplier to the warehouse, and a cyclic delivery schedule from the warehouse to the customer so that the customer demand for each product is satisfied fully at minimum total production, inventory and distribution cost. In the fourth model, we consider a system with one supplier and one customer with a set of orders placed at the beginning of the planning horizon. Unlike the earlier models, here each order can have a different size. Since the shipping capacity per batch is finite, we have to solve an integrated production-distribution scheduling and order-packing problem. Our objective is to minimize the number of delivery batches subject to certain service performance measures such as the average lead time or compliance with deadlines for the orders.