The National Science Foundation established the Decision and Management Sciences Program (DMS) in 1982. Healthy and growing, DMS is likely to affect OR/MS significantly, not only by providing funds for basic research, but also through its vision of a combined theoretical and empirical science of operational and managerial processes, and by its policy of bringing together different disciplines within the program. An NSF workshop held in Dallas in April 1984 sought to identify research opportunities in the decision and management sciences. Promising areas, many of them inviting cross-disciplinary research, appear in the development of measurement-based models for operational processes (both natural and designed and including organizational and managerial activities), choice theory (individual and group choice, values, judgment, and risk behavior), decision support, and the treatment of complexity. Increased knowledge in any of these areas would be likely to have a valuable long-run impact on management practice.

This paper calculates the minimum required frequency between audits of a given type to meet prespecified accuracy goals for a given type of account. Both “100 percent,” as well as sampling type audits are addressed. The effectiveness of a given audit type includes the mean and standard deviation of any residual error that may remain in the account after the audit has been performed and after account balances have been adjusted. The accuracy goals consist of the maximum accumulated error that is considered by management to be tolerable between audits, and the prescribed likelihood that this tolerance level will not be exceeded. The model assumes that certain parameters of the error process have been estimated, but no distributional information is required. Closed form, easy to use, lot-size type formulae are derived which calculate the required frequency (or upper bounds) between audits. The model also provides insights as to the relative cost-effectiveness of various types of audits.

The multivariate exponential smoothing model of Enns, Machak, Spivey and Wrobleski is examined and it is found that its structure is such that it can be estimated by using techniques designed for a univariate exponential smoothing model. Similarly forecasts can be made using algorithms for the univariate model. The model can therefore be handled very easily. A more general univariate time series model, which can include polynomial trends and seasonal factors, is then set up and a multivariate generalisation, analogous to the multivariate exponential smoothing model, is introduced. It is shown that this model can also be handled using algorithms designed for the univariate case.

A paper by Bultez and Naert in the May 1979 issue of Management Science tentatively concludes that profits are relatively insensitive to misspecification of the lag structure of advertising. We examine the conditions under which a firm's profits are most likely to be sensitive to the misspecification of the lag structure. Our analysis indicates how this sensitivity is related to (a) the form of the true lag structure, (b) the demand function, (c) the cost function, and (d) the firm's discount rate.

We consider the problem of determining a schedule of capacity expansions for m producing regions and a schedule of shipments from the regions to n markets so as to meet market demands over a T-period planning horizon at minimum discounted capacity expansion and shipment costs. The proposed algorithm permits capacity expansion costs to be arbitrary nonnegative increasing functions of the expansion amounts, but the shipment (and production) costs are restricted to be proportional to the amounts shipped. The algorithm does not require market demands to be increasing over time. The cost functions are allowed to be nonstationary and the possibility of imports is considered. The proposed heuristic algorithm improves on feasible solutions by simultaneously reassigning several capacity expansions to different regions and/or time periods. A look-ahead feature prevents the algorithm from becoming myopic and a self-learning feature dynamically updates computational parameters. The heuristic algorithm was tested on both randomly generated and real-life based problems with m ≤ 15, and n ≤ 15 and T ≤ 25. The test problems had increasing market demands, capacity expansion costs specified in the form of a concave power function or a fixed charge plus linear function, stationary costs (aside from a constant discount factor), and no imports. Results indicate that for the class of problems tested, the heuristic algorithm is computationally efficient and provides solutions that are closer to optimum than those obtained by previous algorithms.

Data Envelopment Analysis has now been extensively applied in a range of empirical settings to identify relative inefficiencies, and provide targets for improvements. It accomplishes this by developing peer groups for each unit being operated. The use of categorical variables is an important extension which can improve the peer group construction process and incorporate “on-off” characteristics, e.g., presence of drive-in window or not in a banking network. It relaxes the stringent need for factors to display piecewise constant marginal productivities. In so doing, it substantially strengthens the credibility of the insights obtained. The paper treats the cases when the categorical variable can be controllable or uncontrollable by the manager, for the cases of technical and scale inefficiency. The approach is illustrated using real data.

In order to determine an optimum sales strategy for a property it is useful to estimate the distribution of bids which will be received for the property. The more accurate the estimate of the distribution, the higher the expected return will be from following the optimal strategy. This paper presents a technique for estimating the parameters of that distribution based on order statistics. It also describes how a posted list price might affect the sale price data used in this estimation technique, and how the technique can be modified to deal with this type of censored data.

Material Requirements Planning or Manufacturing Resource Planning (MRP) systems originally were designed for a deterministic environment. Often, however, demand for finished products is uncertain and some type of buffering mechanism is necessary to provide desired levels of service. When the schedule is replanned each period, as typically is done, variability of demand may manifest itself in the form of early or “emergency” production runs. Safety stock may serve to avert some or most of these emergency production runs. We have developed a heuristic algorithm which is shown to provide extremely good guidelines for setting safety stock levels.

This paper examines the effect of stochastic task times on the total operating costs of a continuously paced assembly line under the assumption that the line is stopped whenever at least one work station requires more time than allotted. A comprehensive stochastic cost function is integrated into an efficient balancing algorithm to enable an approximately minimum cost balance to be obtained. An experiment was conducted to determine the cost savings resulting from using the stochastic method as compared to two deterministic methods. The stochastic method of this paper produced lower cost balances in most cases studied.

It has been observed that firms alter year-end inventory policies in response to accounting tax incentives. This study proposes a stochastic ordering policy model which quantifies these effects. An innovative feature is its use of two decision variables: an initial order-up-to-level at the beginning of each year and a desired year-end inventory level. The ability to place a second order after demand has been assessed allows for an explicit consideration of the effects of tax incentives on order quantity decisions. Separate formulations are developed for the two most widely-used inventory accounting methods, LIFO and FIFO. The model provides new implications for choices between LIFO and FIFO, the forms of optimal ordering policies under each and for year-end inventory levels. A procedure is given for calculating the optimal FIFO policy. Calculating optimal LIFO policies is very difficult so myopic approximate policies are given which bound the optimal policy.