Increasing manufacturing flexibility is a key strategy for efficiently improving market responsiveness in the face of uncertain future product demand. Process flexibility results from being able to build different types of products in the same plant or production facility at the same time. In Part I of this paper, we develop several principles on the benefits of process flexibility. These principles are that 1) limited flexibility (i.e., each plant builds only a few products), configured in the right way, yields most of the benefits of total flexibility (i.e., each plant builds all products) and 2) limited flexibility has the greatest benefits when configured to chain products and plants together to the greatest extent possible. In Part II, we provide analytic support and justification for these principles. Based on a planning model for assigning production to plants, we demonstrate that, for realistic assumptions on demand uncertainty, limited flexibility configurations (i.e., how products are assigned to plants) have sales benefits that are approximately equivalent to those for total flexibility. Furthermore, from this analysis we develop a simple measure for the flexibility in a given product-plant configuration. Such a measure is desirable because of the complexity of computing expected sales for a given configuration. The measure is ∏(M*), the maximal probability over all groupings or sets of products (M) that there will be unfilled demand for a set of products while simultaneously there is excess capacity at plants building other products. This measure is easily computed and can be used to guide the search for good limited flexibility configurations.

The use of a Visual Interactive Simulation (VIS) model for experimental analysis, where the user initiates runs and gathers information as desired without necessarily any respect for formal analysis, is encouraged by some proponents of VIS and VIS software packages. Proponents of formal output analysis view this approach as dangerous and irresponsible. We designed and executed a laboratory experiment in which 51 subjects solved a case study based around the allocation of trucks in a mining operation in order to investigate the efficacy of VIS to model experimentation. Subjects were provided with a VIS, developed by the authors, which contained a terminating simulation of the system and two different displays: an animation and a dynamically changing histogram. The user could halt execution of the model and change the truck allocation at any time. We found that subjects performed badly relative to a known solution obtained through detailed formal experimentation but performed well compared to solutions they provided prior to use of the model. Use of the animated display was not associated with correct solutions but was associated with more efficient use of the VIS. Subjects who obtained a correct solution investigated fewer alternatives and used fewer interactions than those obtaining incorrect solutions. Finally, we found a significant difference in the process used between subjects providing correct and incorrect solutions.

This article considers the construction of simulation designs for the ordinary least squares estimation of second-order metamodels. Two premises underlie the development of these experimental strategies. First it is assumed that the postulated metamodel may be misspecified due to the true model structure being of third-order. It is therefore important that the locations of the simulation experiments be specified to provide protection against bias, as well as variance, in the estimation of metamodel parameters. The second premise is based on the observation that, in many applications of metamodels, functions of the fitted model coefficients (such as the slope gradients) are of greater interest than the response function. The integrated mean squared error of slopes design criterion that is implemented here addresses both premises. This criterion finds application in various optimum seeking methods and sensitivity analysis procedures. Combinations of four important classes of response surface designs and three pseudorandom number assignment strategies constitute the basis structure of the simulation designs studied. The performance of these simulation designs is evaluated and, subsequently, compared to a similar set of experimental plans that have as their focus the estimation of the response function.

A modeling framework that merges knowledge-based expert systems and decision support systems with management science methods for project evaluation is presented. In particular, the strategic decision to commit to full-scale development of a new product is considered. At the core of the framework are the methods and techniques used for acquiring, modeling and processing the expert knowledge and data. Methods and techniques used include scoring models, logic tables, the analytic hierarchy process, discriminant analysis, and rule-based systems. The suggested modeling approach obtains the benefits of normative modeling as well as the flexibility and developmental advantages of expert systems. Additional benefits include reduced information processing and gathering time, which can help to accelerate the product development cycle. Potential spin-offs of this research include applications for project evaluation throughout the product development cycle and other areas such as capital budgeting. Finally, a series of related case studies that have successfully implemented this framework is described.

This paper presents a model of a bottleneck facility that performs two distinct types of operations: “regular” and “repair.” Both switch-over time and cost are incurred when the facility switches from performing one type of operation to a different type. Upon the completion of a batch of jobs in the regular mode, each batch is subjected to a test, where the entire batch (of jobs) will be classified accordingly as either nondefective, repairable, or nonrepairable. A nondefective batch continues its process downstream, a nonrepairable batch is scrapped, and a repairable batch can be cycled back to the bottleneck facility for repair. The objective of this paper is to determine the optimal repair policy for the bottleneck facility so that the long run average operating profit is maximized. We first characterize the optimal repair policy by showing that the optimal repair policy must take one of the two forms: a “repair-none” policy under which all repairable batches are scrapped, or a “repair-all” policy under which all repairable batches are repaired. We then develop optimality conditions for the repair-none policy and the repair-all policy. When the repair-all policy is optimal, we further show that there exists an optimal “threshold” operating policy that can be described as follows: upon completion of a regular batch, switch over to the repair mode only if the number of available repairable batches exceeds a certain threshold value. We also evaluate the impact of batch sizes, yield, and switch-over cost on the optimal operating policy.

A model for efficiency evaluation based upon the theory of chance constrained programming is developed. The model uses a piecewise linear envelopment of confidence regions for observed stochastic multiple-input multiple-output combinations in the Data Envelopment Analysis (DEA) tradition. The model allows for an exogenous decomposition of the total variation in data for each Decision Making Unit (DMU). By varying certain probability levels the model can provide estimates of the sensitivity of efficiency scores regarding an unknown amount of noise in date. An application of the method in an evaluation of the research activities in economic departments at Danish Universities is presented.

Path dependent securities depend on current and past values of underlying state variables. Consequently, the usual backward evaluation technique is difficult to apply since state variable values existing earlier in real time are unknown. This paper develops a series of propositions which makes possible the pricing of a certain class of both American and European versions of these path dependent securities.

The literature in the field of interior point methods for Linear Programming has been almost exclusively algorithmic oriented. Very few contributions have been made towards the theory of Linear Programming itself. In particular none of them offer a simple, self-contained introduction to the theory of Linear Programming and linear inequalities. The purpose of this paper is to show that the interior point methodology can be used to introduce the field of Linear Programming. Starting from scratch, and using only elementary results from calculus and linear algebra, we prove that for every value of the barrier parameter, the logarithmic barrier function for the primal-dual problem has a unique minimizer, and that the path of these minimizers (the central path) converges to a strictly complementary pair of optimal solutions. These results were proved more than a decade ago with advanced mathematical arguments. Our proofs are new: they are also simpler and often more natural than the ones currently known. They provide a new approach to the fundamental results of Linear Programming, including the existence of a strictly complementary solution, and the strong duality theorem.