AN EMPIRICAL EVALUATION OF INDIVIDUAL ITEM FORECASTING MODELS

A computer simulation experiment was replicated to correct errors in an earlier paper and to compare seven individual item forecasting models across five different demand patterns. Results confirm previous findings that the better forecasting model depends upon the demand pattern and the forecast horizon, as well as the noise level. Nevertheless, exponential double smoothing emerged as the most robust model.     

THE CLASSIC ECONOMIC PRODUCTION QUANTITY MODEL WITH SETUP COST AS A FUNCTION OF CAPITAL EXPENDITURE*

The formulation of the classic economic production quantity (EPQ) model is extended to include setup cost as a function of capital expense. Additional capital will buy reduced setup cost. Thus, the objective now is to balance holding, setup, and capital expenses. This new formulation is solved under conditions where setup cost varies exponentially and linearly as a function of capital expense. Decision rules are formulated to indicate under what conditions setup cost reduction reduces total cost. For the linear function, it is shown that once the decision to reduce setup cost is justified, the optimal choice is the minimum setup cost that is technologically feasible.     

Concepts,Theory, and Techniques THE LOT-SIZE REORDER-POINT MODEL WITH UPSTREAM-DOWNSTREAM UNCERTAINTY*

One assumption in the classical lot-size problem is certainty in the amount requisitioned. In many practical situations, however, the amount received may be a random variable. In this paper, we discuss a lot-size inventory problem in which the quantity received does not necessarily match the quantity requisitioned. We develop explicit and approximate solutions for the back-orders case by assuming that the standard deviation of the amount received is linearly related to the quantity requisitioned.     

SOCIAL LEARNING THEORY AND THE IMPLEMENTATION OF PRODUCTION INNOVATION*

In a study of the implementation of production innovation using a social learning theory focus, four organizations were monitored with a real-time study design. The following propositions were strongly supported: (1) During the initial implementation period (3–4 months), the stress felt by personnel in new user organizations was significantly higher (p < .01) than for organizations previously experienced with technology; (2) Firms experienced with the new technology obtain early success when compared with inexperienced firms (p < .01). Moderate support was found for the proposition that new users rely heavily on original equipment manufacturers for demonstration to acquire new skills, but gradually models within the user organization emerge as the dominant influence. The proposition that new users have significantly higher imitation scores than users experienced with the technology was only weakly supported. No substantive or significant changes in the attitudes of users toward the innovation were found during the first three to four months of the implementation period. These and other results suggest that longitudinal studies of this type ought to be carried out over periods in excess of 2 1/2 years. Implications for the management of change are presented.     

PROCEDURES FOR DETERMINING RELATIVE FREQUENCIES OF PRODUCTION/ORDER IN MULTISTAGE ASSEMBLY SYSTEMS*

We show that the problems of determination lot sizes in a multistage assembly system for the case of instantaneous production and constant demand for the end item can be reduced to the problems of determining relative frequenceis of production/order for items at each production stage. We further show that such frequencies are independent of the demand levels. Optimal and near-optimal solution procedures for this reduced problem are provided. The near-optimal procedure successively treats each stage of production as a final production stage while simulatenously incorporating decisions made at lower stages into decisions made at higher stages. Experimental results show that the near-optimal procedure results in optimal solutions 75 percent of the time and performs considerably better than representative heuristics available in the literature. Further, its performance is relatively less susceptible to product/structural characteristics of the system.     

THE PERFORMANCE OF A SIMPLE INCREMENTAL LOT-SIZING RULE IN A MULTILEVEL INVENTORY ENVIRONMENT

A simple incremental cost approach to lot sizing was tested in a multilevel inventory environment. The incremental approach has not previously been tested in a large-scale study involving multiple product-structure levels. Using the Wagner-Whitin (WW) algorithm as a benchmark, the simple incremental rule (IPPA) was compared to three heuristic procedures (LFL, EOQ, and POQ) frequently used in material requirements planning (MRP) lot-sizing research. The incremental rule consistently generated lower total order/setup and carrying costs than the three heuristics across the 3,200 multilevel test situations examined. In many of the test situations, the incremental rule also outperformed the WW benchmark.     

THE TIMING OF LABOR TRANSFERS IN DUAL RESOURCE-CONSTRAINED SYSTEMS: “PUSH” VS. “PULL” RULES*

Altough the dual resource-constrained (DRC) system has been studied, the decision rule used to determine when workers are eligible for transfer largely has been ignored. Some earlier studies examined the impact of this rule [5] [12] [15] but did not include labor-transfer times in their models. Gunther [6] incorporated labour-transfer times into his model, but the model involved only one worker and two machines. No previous study has examined decision rules that initiate labor transfers based on labor needs (“pull” rules). Labor transfers always have been initiated based on lack of need (“push” rules). This study examines three “pull” variations of the “When” labor-assignment decision rule. It compares their performances to the performances of two “push” rules and a comparable machine-limited system. A nonparametric statistical test, Jonckheere's S statistic, is used to test for significance of the rankings of the rules: a robust parametric multiple-comparison statistical test, Tukey's B statistic, is used to test the differences. One “pull” and one “push” decision rule provide similar performances and top the rankings consistently. Decision rules for determining when labor should be transferred from one work area to another are valuable aids for managers. This especially is true for the ever-increasing number of managers operating in organizations that recognize the benefits of a cross-trained work force. Recently there has been much interest in cross-training workers, perhaps because one of the mechanisms used in just-in-time systems to handle unbalanced work loads is to have cross-trained workers who can be shifted as demand dictates [8]. If management is to take full advantage of a cross-trained work force, it need to know when to transfer workers.     

A COMPARISON OF LEARNING CURVE ANALYSIS AND MOVING AVERAGE RATIO ANALYSIS FOR DETAILED OPERATIONAL PLANNING*

This study tests the use of learning curve analysis for production planning at the detailed component level under various conditions, represented by factors of product turnover rate, learning rate, variance levels, and planning horizon length. It also presents an alternative to learning curve analysis that considers aggregation of cost data across time. This alternative is periodic revision of standard cost data using moving average forecasts to reflect productivity trends. Results of this study indicate that in most circumstances a moving average analysis can provide better estimates of short-term, detailed component operations costs than either a learning curve analysis or a standard analysis.     

Scheduling with Sequencing Flexibility*

This study examines the effects of sequencing flexibility on the performance of rules used to schedule operations in manufacturing systems. The findings show that taking advantage of even low levels of sequencing flexibility in the set of operations required to do a job results in substantial improvement in the performance of scheduling rules with respect to mean flowtime. Differences in the mean flowtime measure for various rules also diminish significantly with increasing sequencing flexibility. Performance improvements additionally result for such due-date related performance measures as mean tardiness and the proportion of jobs tardy. At high levels of sequencing flexibility, some nonparametric scheduling rules outperform the shortest processing time rule in terms of the mean flowtime criterion. Rules based on job due dates also outperform rules based on operation milestones in terms of tardiness related criteria at high levels of sequencing flexibility. The implications of these findings for the design of manufacturing systems and product design are noted.     

Vendor Selection with Bundling*

Product bundling has become increasingly prevalent not only in consumer goods but also in the industrial sector. We study a purchasing problem in which a buyer must obtain necessary numbers of various stock items from a variety of vendors who charge different prices, have limited capacities and different levels of quality, and offer bundled products at discounted prices. We examine relationships among different bundling scenarios and show that the most general scenario is one in which free items are given to the buyer when sufficient quantities are purchased. We develop a mixed integer linear program that finds the purchasing strategy for the buyer that minimizes the total purchase cost. We present computational results which indicate that the problem is very tractable to solve optimally on a personal computer with standard optimization software. Finally, three extensions of the model are discussed.     

Effects of Product Structure Complexity on Multi-level Lot Sizing*

As an input to the materials requirement planning (MRP) process, the product structure interacts with the lot-sizing rules to affect inventories, materials flow, and production costs. Despite engineering constraints, considerable latitude is still available to construct alternate product structures for the same product. An important concern of MRP managers and designers is the impact of product structure complexity on the cost performance of the lot-sizing rules. To date, there exists no detailed research that provides managers with guidelines that relate the lot-sizing rules to individual product structure parameters. We report on an extensive experiment to test the effects of individual product structure parameters on the relative cost performance of 11 lot-sizing rules. Three parameters—the number of items, number of levels, and commonality index—are proposed to characterise product structure complexity and used as factors in an experiment involving a large variety of product structures. The results indicate that all three parameters affect the relative cost differences but not the ranking of the rules. The overall best lot-sizing rule is Bookbinder and Koch's [11] rule.     

Manufacturing Proactiveness and Performance*

Since Skinner's [40] landmark article depicting the manufacturing function as the “missing link” in corporate strategic processes, a portion of the blame for inferior performance in many firms has been attributed to the subordinate strategic position of manufacturing. It has been argued that part of the solution to misalignments between the capabilities possessed by manufacturing and the requirements dictated by customers is for manufacturing to take a more proactive stance. However, little research has been reported which examines manufacturing proactiveness empirically. In this paper, we address this gap by developing an operational definition of manufacturing proactiveness and testing empirically whether a link exists between proactiveness and performance based on data collected from a sample of manufacturers. Based on the manufacturing strategy literature, we identify two major dimensions of manufacturing proactiveness: (1) the degree of manufacturing's involvement in the strategic processes of the business unit; and (2) the degree of commitment to a long-term program of investments in manufacturing structure and infrastructure aimed at building capabilities in anticipation of their need. We develop reliable scales for measuring each of the dimensions of proactiveness and use the data to provide evidence of a clear link between manufacturing proactiveness and business performance. We show that investments in structural programs coupled with either high levels of manufacturing involvement in strategic processes or planned investments in infrastructural programs correlate with higher than average performance.     

Distinguishing Features of Just-in-Time Systems in the Make-to-Order/Assemble-to-Order Environment*

Few studies of just-in-time (JIT) implementation examine their significance in make-to-order manufacturing environments. This study examines the relative importance of several operating variables that are characteristic features of JIT systems within such environments in North America. The results suggest that the most salient features of JIT for make-to-order and assemble-to-order firms are the elimination of waste (in the form of time and defects), reduced setup time, reduced lotsize, and a smaller pool of suppliers. In addition, the results suggest that the use of non-domestic suppliers can possibly hinder efforts at JIT implementation of materials procurement.     

Standard Container Size Discounts and the Single-Period Inventory Problem*

This paper analyzes the problem of choosing the optimal order quantity and its associated number of standard containers making up the order for single-period inventory models under standard container size discounts. A range is determined that contains the optimal order quantity. Two algorithms are presented. The first algorithm solves the general case in which there is no restriction on the types of containers included in an order. The second algorithm solves a more restricted policy that requires the buyer to accept the order with successively smaller container sizes.     

ON THE IMPLICATIONS OF SPECIFICATION UNCERTAINTY IN FORECASTING*

Forecasters typically select a statistical forecasting model from among a set of alternative models. Subsequently, forecasts are generated with the chosen model and reported to management (forecast consumers) as if specification uncertainty did not exist (i.e., as if the chosen model were the “true” model of the forecast variable). In this note, a well-known Bayesian model-comparison procedure is used to illustrate some of the ambiguities and distortions of forecasts that do not reflect specification uncertainty. It is shown that a single selected forecasting model (however chosen) will generally misstate measures of forecast risk and lead to point and interval forecasts that are misplaced from a decision-theoretic point of view.     

UNCERTAINTY AND THE LOT-SIZE REORDER-POINT MODEL

Noori and Keller [1] recently proposed a lot-size reorder-point model in which the quantity received is a random variable. The authors failed, however, to explore the full implications of their initial definitions. This oversight leads to some misleading inferences that we try to correct in this note.     

Sequencing Rules for Productivity Improvements in Underground Coal Mining*

In this paper, we report on a unique two-machine, dynamic sequencing problem that was involved in a project designed to improve the productivity of underground coal mining operations. It was discovered that mine output was dependent on the particular sequence of faces the continuous miner and bolter worked. The problem is characterized by two unique features. First, the problem is unique because of the geological and safety constraints that must be satisfied by the sequence. Second, the scheduling problem is dynamic because the available job sets change as work progresses, and the production times change as coal is mined. Three heuristic approaches were developed and tested using mine parameters. The best heuristic had a worst case scenario of being within 2.7 percent of the optimal sequence and only a 1.1 percent chance of determining a better sequence. This heuristic resulted in an 8.1 percent productivity improvement.     

A Simulation Study of Sequencing Rules in a Kanban-Controlled Flow Shop*

This paper addresses the problem of sequencing in decentralized kanban-controlled flow shops. The kanban production control system considered uses two card types and a constant withdrawal period. The flow shops are decentralized in the sense that sequencing decisions are made at the local workstation level rather than by a centralized scheduling system. Further, there are no material requirements planning (MRP)-generated due dates available to drive dispatching rules such as earliest due date, slack, and critical ratio. Local sequencing rules suitable for the decentralized kanban production-control environment are proposed and tested in a simulation experiment. These rules are designed so that they can be implemented with only the information available at the workstation level. Example sequencing problems are used to show why the shortest processing time rule minimizes neither average work-in-process inventory nor average finished-goods withdrawal kanban waiting time. Further, it is shown how work station supervisors can use the withdrawal period, in addition to the number of kanbans, to manage work-in-process inventories.     

MODELS OF PARTS-ORIENTING SYSTEMS*

An analytical model to estimate, evaluate, and compare the behavior and performance of parts-orienting systems for automated manufacturing systems is developed. The model can be applied to parts-orienting systems for both traditional series-structured manufacturing systems and nontraditionally structured flexible manufacturing systems. Numerical examples are given for typical system configurations such as the pure series system, the pure parallel system, and the pure feedback system.