David P. Kopcso

Learning curves and lot sizing for independent and dependent demand

Abstract This paper explores the effect of learning curve cost behavior, as opposed to linear, on lot sizing. The first portion of the paper develops optimizing models for the independent demand situation. The second portion examines lot sizing for dependent demand, developing a lot sizing rule similar to Part Period Balancing. After examining the shortcomings of previous attempts at the independent demand lot sizing problem, two models are derived. Excluding material costs (for an assembly operation, the cost of all components), the optimal lot size is seen to vary linearly with demand and inversely with the carrying cost rate. When material costs are included a smaller optimal lot size is derived. The difference between the two, expressed as a fraction of the smaller lot size, equals the material/labor ratio of the last unit produced in the smaller lot size. For dependent demand, the incremental model developed by Freeland and Colley as an improvement on Part Period Balancing is used as a beginning concept. An analogous model, called Assembly Period Balancing, is developed for learning curve cost behavior. The decision rule for combining lots is expressed as a comparison of the material/labor ratio of the lot considered for combining with another expression involving the carrying cost rate, relative lot size and the learning curve exponent. Finally, cost data from an electronics manufacturer are used to examine the cost penalties of failing to recognize learning curve cost behavior. It is shown that optimal lot sizes for learning curve costs can be much larger than those obtained assuming linear costs. It is also shown that much larger lots can be economically combined in the dependent demand case when costs follow a learning curve.

Classifying the uncertainty arithmetic of individuals using competitive learning neural networks

Abstract The application of artificial neural network technology to a host of problems in pattern recognition has long been advocated. Several analyses comparing the performance of neural networks to the standard methods for achieving machine classification and machine learning, such as statistical pattern recognition and ID3, have been reported. Typically, supervised learning has been used and the specific learning algorithm has been back propagation. For many classification type problems, a priori categories are not available, that is, one does not know explicitly the number of categories existent nor the boundaries delineating these categories. Therefore, known targets with which to train the network are not available. A supervised learning approach is not appropriate under these circumstances; an unsupervised learning algorithm is required. In this article we report on the use of an unsupervised competitive learning algorithm as a classifier. The network was used to classify individuals into categories based on differences in the manner in which individuals manipulate the uncertainty associated with the chaining of rules. The experiment, from which the data to be classified were obtained, is described, results of the neural network approach are compared to classification using a distance measure and to classification using a standard clustering algorithm.

Logarithmic Transformations in Regression: Do You Transform Back Correctly?

Abstract The logarithmic transformation is often used in regression analysis for a variety of purposes such as the linearization of a nonlinear relationship between two or more variables. We have noticed that when this transformation is applied to the response variable, the computation of the point estimate of the conditional mean of the original response variable is often incorrect. Although the correct procedure has long been known in the scientific community and is well documented, an incorrect or misleading procedure is used in many business statistics textbooks. This incorrect procedure results in errors that can be quite significant. Our article uses a real-data business example which, in the context of making a decision about an advertising charge for a magazine, illustrates the correct procedure. This example also provides a sense of the magnitude of the error that would result if the incorrect procedure were used. The six percent error in our example could be substantially higher for other applications.

A comparison of the manipulation of certainty factors by individuals and expert system shells

The treatment of uncertainty in expert system shells is addressed, starting with a review of the modeling of uncertainty by expert system shells. An experiment to replicate earlier work investigating the manner in which individuals manipulate certainty factors in comparison to commercial shells is discussed. Comparisons are made among seven commercial shells, both personal-computer (PC)-based and mainframe-based, and individuals. A significant difference between individuals and shells themselves is indicated. Some implications for both expert system and decision-support-system methodologies are discussed.<<ETX>>

The application of artificial neural networks to quality control charts

Reports on the development of artificial neural networks that function as alternatives to conventional quality control charts. Multilayered feedforward networks using a backpropagation learning algorithm were trained and tested. The results illustrate the feasibility of using artificial neural networks to detect out-of-tolerance conditions in a manufacturing process.<<ETX>>

The processing of numerical uncertainty associated with the components of if then rules: experiments with human subjects

The authors present the results of an experiment that was conducted to explore how individuals interpret and manipulate uncertainty in basic logical operations. Results indicate that the commonly used models are not as universally appropriate as has been assumed. These results have implications for the process of knowledge acquisition, the design of the system/user interface, and general issues of system development. Collecting information to construct an expert system, duplicating reasoning by using an expert system, and displaying results to decision makers require an understanding of how people actually reason with and use uncertain information.<<ETX>>

Using Simulation to Model Customer Behavior in the Context of Customer Lifetime Value Estimation

This article illustrates how simulation can be used in the classroom for modeling customer behavior in the context of customer lifetime value estimation. Operations research instructors could use this exercise to introduce multiperiod spreadsheet simulation models in a business setting that is of great importance in practice, and the simulation approach to teaching this subject could be of interest also to marketing and accounting instructors. At Babson College, the spreadsheet simulation exercise is part of an integrated one-case teaching day of the marketing, accounting, and operations research disciplines in the full-time MBA program, but the exercise is directly transferable to stand-alone courses as well. In our experience, students have felt empowered by the ability to incorporate their ideas about customer behavior directly into customer lifetime value models, and have appreciated the ease with which simulation enables them to obtain intuition about the sensitivity of their estimates to different assumptions.

Artificial Neural Networks as Alternatives to Statistical Quality Control Charts in Manufacturing Processes

The concept of quality has reentered the vocabulary of American business. The perception, whether founded in reality or not, that American products are inferior to their foreign counterparts, has contributed to the competitive disadvantage now faced by many American firms.

Case Article—Idiopathic Pulmonary Fibrosis

This article describes a case in which decisions are made by two biopharmaceutical firms in pursuit of FDA approval of a drug to treat idiopathic pulmonary fibrosis (IPF). The case contains information on each firm’s estimates of costs, revenues and likelihood of success, as well as average values for these available from the scientific literature. The case provides an opportunity to apply decision analysis in the form of decision trees to various decision problems and to perform sensitivity analysis. It can be used as an introduction or as an application of decision trees after an introduction. The students are first exposed to one firm’s simple decision based on expert opinion, which is then modified by the inclusion of data. The advantage of expressing information in a tree diagram becomes apparent. The tree diagram is then examined to expose hedging strategies, one of which introduces the second larger firm as a potential licensee. The second firm presents its own view of the decision process based on its own expertise, thus allowing for a rich discussion of sensitivity analysis. Students are to evaluate the first firm’s approach to decision making and whether the second firm should be a licensee or not. Teaching Note: Interested Instructors please see the Instructor Materials page for access to the restricted materials. To maintain the integrity and usefulness of cases published in ITE , unapproved distribution of the case teaching notes and other restricted materials to any other party is prohibited.

Case—Idiopathic Pulmonary Fibrosis

The Executive VP of Scientific and Medical Affairs looked up at the clock on the wall; it read 5:38 p.m. Another glance, down this time, revealed a stack of empty coffee cups and a pile of peppermint candy wrappers decorating the coffee table. He and InterMune’s CFO had been locked up in the VP’s office since the morning discussing the next strategic move for the company to expand its pulmonary drug portfolio. InterMune was founded in 1998 by W. Scott Harkonen, M.D., as InterMune Pharmaceuticals, Inc. Originally a wholly-owned subsidiary of Connetics Corporation in Burlingame, California, the company was reincorporated at the time it went public in 2000 as InterMune, Inc.1 InterMune’s initial intent was to develop pharmaceutical products to treat a wide range of pulmonary and infectious diseases such as cystic fibrosis, pulmonary fibrosis, tuberculosis and hepatitis C, as well as certain cancers, such as ovarian cancer. InterMune’s business model was to license existing drugs and proteins from large, established pharmaceutical suppliers and to expand the use of those compounds into new therapeutic areas through traditional clinical development activities. After the first few years, InterMune’s management found that reorganization was needed—the company had to narrow its focus if it was to achieve profitability. Approved products were divested and clinical trials abandoned in order to focus the company’s development activities specifically on pulmonology