Applications of Management Science: In Productivity, Finance, and Operations: Volume 12

This peer-reviewed volume is part of an annual series, dedicated to the presentation and discussion of state-of-the-art studies in the application of management science to the solution of significant managerial decision-making problems. It is hoped that this research annual will significantly aid in the dissemination of actual applications of management science in both the public and private sectors. Volume 1 is directed toward the applications of mathematical programming to (1) multi-criteria decision making, (2) supply chain management, (3) performance management, and (4) risk analysis. Its use can be found both in the university classes in management science and operations research (management and engineering schools), as well as to both the researcher and practitioner of management science and operations research. Series information at: http://www.elsevier.com/locate/series/mansc

Data envelopment analysis is used here to track organizational efficiency over time. Input and output values are discounted with the consumer price index for equity. The model obtains annual comparisons of efficiency by obtaining outputs from a designated set of inputs.

This research deals with the evaluation of the efficiency of consumer freight package delivery. Model inputs include number of delivery and administrative employees, labor hours, operating costs, and number of deliver vehicles. Outputs include number of packages delivered, percent on-time, percent lost, percent damaged, revenue per package, and customer satisfaction. The methodology used to evaluate the efficiency of the decision-making units (DMUs) under consideration is data envelopment analysis (DEA) involving multiple criteria. The inclusion of additional criteria beyond basic DEA efficiency can improve discriminating power between DMUs and also tends to yield more reasonable weights on model inputs and outputs.

Data Envelopment Analysis (DEA) is a nonparametric mathematical programming technique used to measure the relative efficiency of the production organization's operations. This paper presents the theoretical measures of the railway systems, along with the bootstrap DEA analysis. A DEA model is applied to evaluate the relative efficiency of railway operations of 29 UIC (Union Internationale des Chemins de fer) countries, based on the data obtained from the International UIC publications. The bootstrap DEA analysis provides information (bias estimates) on the sensitivity of the DEA efficiency index to the sampling variations. The model results are analyzed and evaluated in terms of their relative operational performance efficiency. The model results facilitate an organization's decision-making by providing valuable information.

Researchers and corporate managers are interested in firm performance and its measurement. Data envelopment analysis (DEA) offers a powerful analytical tool that can be utilized to identify and measure firm performance. This paper provides a comprehensive set of key input and output variables necessary for DEA analysis of electric utility performance.

A survey of multi-objective scheduling techniques on the job shop problem is offered in this chapter. The survey traces the development of techniques from Integer programming to genetic algorithms that take advantage of the power of recent computing technology. Applications are in areas as diverse as job scheduling, nurse scheduling, and groundwater monitoring.

This paper will detail the development of a multi-objective mathematical programming model for audit sampling of balances for accounts receivable. The nonlinear nature of the model structure will require the use of a nonlinear solution algorithm, such as the GRG or the genetic algorithm embedded in a Solver spreadsheet modeling system, to obtain appropriate results.

Frequently, problems involving a management activity involve the incurring of a setup charge for the activity. In these cases, the total cost of the activity is the sum of the variable cost related to the level of the activity and a set up cost required to initiate the activity. In this research, we will also use goals that involve demand management of service centers based upon the revenue potential of the customer districts they will serve.

Pharmaceutical companies are faced with identifying development compounds for their Drug Development Processes (DDPs) that will not only gain approval for sale by the regulatory agencies, such as the Food and Drug Administration (FDA), but also establish a sustainable and profitable market presence. This identification of compounds for the DDP includes projection of objective criteria, such as ability to generate revenue and profitability (Financial) and safety and efficacy (Clinical), as well as more subjective criteria, such as determination of insurance coverage by payers, such as the Centers for Medicare and Medicaid Services and pricing (Reimbursement), ability to produce a product of consistent quality (Manufacturing), and attain approval for sale in a timely manner (Registration). The Analytical Hierarchy Process (AHP) is a multi-criteria decision model that can integrate both objective and subjective information. This study applies the AHP methodology to the identification of compounds resulting in a dynamic application of the model that can be used by pharmaceutical companies to determine the best compounds to put in the DDP, at a time when the cost of conducting clinical evaluations for development compounds is very high and global market conditions are evolving.

In this paper, we consider the disassembly-to-order (DTO) problem, where a variety of returned products are disassembled to fulfill the demand for specified numbers of components and materials. The objective is to determine the optimal numbers of returned products to disassemble so as to maximize profit and minimize costs. We model the DTO problem using a multi-criteria decision-making approach. Since the conditions of returned products are unknown, the yields from disassembly are considered to be stochastic. To solve the stochastic problem, we use one of the two heuristic approaches (viz., one-to-one approach or one-to-many approach) that converts the problem into a deterministic equivalent. We compare the performance of the two heuristic approaches using a case example.

We address the problem of selecting a topological design for a network having a single traffic source and uncertain demand at the remaining nodes. Solving the associated fixed charge network flow (FCF) problem requires finding a network design that limits both the fixed costs of establishing links and the variable costs of sending flow to the destinations. In this paper, we discuss how to obtain a sequence of optimal solutions that arise as the demand intensity varies from low levels to high. One of the network design alternatives associated with these solutions will be chosen based upon the dominant selection criteria of the decision maker. We consider both probabilistic and non-probabilistic criteria and compare the network designs associated with each. We show that the entire sequence of optimal solutions can be identified with little more effort than solving a single FCF problem instance. We also provide solution approaches that are relatively efficient and suggest good design alternatives based upon approximations to the optimal sequence.

Truckload routing has always been a challenge. This paper explores the development of continuous flow truckload routes, which resemble less than truckload routes, and a new way to formulate the truckload routing problem (TRP). Rather than view the problem as a succession of origin/destination pairs, we look at the problem as a series of routing triplets. This enables us to use alternate solution methods, which may result in greater efficiency and improved solutions.

Recent market structure reviews have shown a shift of retailing power from manufacturers to retailers. Retailers have equal or even greater power than a manufacturer when it comes to retailing. Based on this new market phenomenon, we intend to explore the role of vertical cooperative (co-op) advertising with respect to transactions between a manufacturer and a retailer. In this paper, we explore the role of vertical co-op advertising efficiency of transactions between a manufacturer and a retailer. We address the impact of brand name investments, local advertising, and sharing policy on co-op advertising programs in a manufacturer–retailer supply chain. Game theory concepts form the foundation for the analysis. We begin with the classical co-op advertising model where the manufacturer, as the leader, first specifies its strategy. The retailer, as the follower, then decides on its decision. We then relax the assumption of retailer's inability to influence the manufacturer's decisions and discuss full coordination between the manufacturer and the retailer on co-op advertising.

Disassembly takes place in remanufacturing, recycling, and disposal, with a line being the best choice for automation. The disassembly line balancing problem seeks a sequence that is feasible, minimizes the number of workstations, and ensures similar idle times, as well as other end-of-life specific concerns. Finding the optimal balance is computationally intensive due to exponential growth. Combinatorial optimization methods hold promise for providing solutions to the problem, which is proven here to be NP-hard. Stochastic (genetic algorithm) and deterministic (greedy/hill-climbing hybrid heuristic) methods are presented and compared. Numerical results are obtained using a recent electronic product case study.

“Mirror Worlds” were suggested by David Gelernter based on a bold assertion: “You will look into a computer screen and see reality.” With mirror worlds, managers could be proactive, anticipating what might happen and acting accordingly, instead of waiting till events happen and then reacting. This paper extends the notion of mirrors worlds to supply chain management. In the case of supply chain management, managers could test the impact of making changes in their supply chains to study the impact.

However, mirror worlds could be extended to help actually monitor and manage supply chains to respond and adapt to changes in the world that affected the supply chain. In particular, mirror worlds could be “real” worlds if control for some of the activities between supply chain participants is in effect “turned over” to the mirror world. In that case, the mirror world would show the actual world, with the system making many of the decisions.

There has been an increasing amount of research on personnel selection in many business disciplines (Hough & Oswald, 2000; Breaugh & Starke, 2000). Research on internal auditor selection, however, has had limited exposure in the auditing literature (Bailey, Gramling, & Ramamoorti, 2003). Recently, Seol and Sarkis (2005) introduced an analytic hierarchy process (AHP) model that used a decision hierarchy based on the CFIA (competency framework for internal auditing) framework. A limitation of AHP, however, is the assumption of strict hierarchical relationship that needs to exist among factors.

The purpose of this paper is an introduction of a more robust model, the analytical network process (ANP), which relaxes the strict hierarchical and decomposition levels of the hierarchy and incorporates possible interrelationships and interdependencies of various personnel selection criteria, factors, and alternatives. In illustrating the application, we return to the CFIA model framework, describe how and where interdependencies exist amongst the CFIA factors/attributes, and how ANP is used in the internal auditor selection process. The illustration will also describe some sensitivity analysis for the ANP approach. The tool is not without its limitations that include the potential for geometrically more questions and information elicitation from the decision makers. Finally managerial and research implications associated with the technique and results are described.

A warranty policy involving two attributes, for example time and usage, is considered. Usage is assumed to be related to time through the usage rate, which is considered to be a random variable satisfying a specified probability distribution. The paper analyzes a policy where warranty is not renewed on product failure, within the specified time period and amount of usage, but is minimally repaired. Unit cost of minimal repair, conditional on the usage rate, is assumed to be a non-linear function of the two warranty parameters. Expressions for the expected warranty costs per unit sales are derived. Applications of the results are presented through sample computations. The results demonstrate the use of warranty cost information in selecting the parameters of the warranty policy.

Using belief functions, this paper develops a model of the situation of a management team trying to decide if a cost process is in control, or out of control and, thus, in need of investigation. Belief functions allow accounting for uncertainty and information about the cost processes, extending traditional probability theory approaches. The purpose of this paper is to build and investigate the ramifications of that model. In addition, an example is used to illustrate the process.

This paper provides a theoretical framework for application of Chance-Constrained Programming (CCP) in situations where the coefficient matrix is random and its elements are not normally distributed. Much of the CCP literature proceeds to derive deterministic equivalent in computationally implementable form on the assumption of “normality”. However, in many applications, such as air pollution control, right skewed distributions are more likely to occur. Two types of models are considered in this paper. One assumes an exponential distribution of matrix coefficients, and another one uses an empirical approach. In case of exponential distributions, it is possible to derive exact “deterministic” equivalent to the chance-constrained program. Each row of the coefficient matrix is assumed to consist of independent, exponentially distributed random variables and a simple example illustrates the complexities associated with finding a numerical solution to the associated deterministic equivalent. In our empirical approach, on the other hand, simulated data typically encountered in air pollution control are provided, and the data-driven (empirical) solution to the implicit form of deterministic equivalent is obtained. Post-optimality analyses on model results are performed and risk implications of these decisions are discussed. Conclusions are drawn and directions for future research are indicated.

Based on a continuous version of the Lanchester advertising model for a duopoly, a mathematical model is developed to determine the optimal advertising policy of a firm responding to the advertising pulsation policy of its competitor. A Dynamic Programming (DP) approach has been employed to arrive at the optimal solution.

It has been mainly demonstrated that under a concave or linear advertising response function, the focal firm's DP policy is superior to its Uniform Advertising Policy (UAP) counterpart (constant advertising spending over time), irrespective of the advertising pulsation policy employed by its rival. Under a convex advertising response function, on the other hand, the focal firm's DP policy is superior to its Advertising/Maintenance Pulsing Policy (APMP) and Advertising Pulsing Policy (APP) counterparts (alternating advertising spending at two levels), irrespective of the advertising pulsation policy used by the competitor.