Fred F. Easton

Pricing and lead time decisions for make-to-order firms with contingent orders

Abstract Make-to-order (MTO) firms have few standard products and volatile, difficult-to-predict demand. A production order usually stems from a successful bid, which presents the MTO firms terms (including price and lead time) to satisfy a prospective customers stated requirements. The customer may decide to accept, reject, or modify these terms. Until the customer decides, the tendered bid is a contingent demand on the MTO firms future production capacity. In the short run, this capacity may be relatively fixed and, if the customer awards the contract to another bidder, it may simply go to waste. To hedge against this possibility, the MTO firm can bid on other projects that require the same capacity. However, this strategy introduces a new source of lead time uncertainty. If two or more booked orders must contend for the same resources, some work will inevitably be “bumped” to later time periods. Thus, a hedging strategy carries the risk of penalties for late deliveries. In this paper we model this little-discussed source of lead time uncertainty and introduce a technique that simultaneously optimizes pricing and lead time decisions for MTO firms with contingent orders. We illustrate the procedure with a simple numerical example.

A distributed genetic algorithm for deterministic and stochastic labor scheduling problems

Abstract A recurring operational decision in many service organizations is determining the number of employees, and their work schedules, that minimize labor expenses and expected opportunity costs. These decisions have been modeled as generalized set covering (GSC) problems, deterministic goal programs (DGP), and stochastic goal programs (SGP); each a challenging optimization problem. The pervasiveness and economic significance of these three problems has motivated ongoing development and refinement of heuristic solution procedures. In this paper we present a unified formulation for these three labor scheduling problems and introduce a distributed genetic algorithm (DGA) that solves each of them. Our distributed genetic algorithm operates in parallel on a network of message-passing workstations. Separate subpopulations of solutions evolve independently on each processor but occasionally, the fittest solutions migrate over the network to join neighboring subpopulations. With its standard genetic operators, DGA frequently produces infeasible offspring. A few of these are repaired before they enter the population. However, most enter the population as-is, carrying an appropriate fitness penalty. This allows DGA to exploit potentially favorable adaptations that might be present in infeasible solutions while orienting the locus of the search near the feasible region. We applied the DGA to suites of published test problems for GSC, DGP, and SGP formulations and compared its performance with alternative solution procedures, including other metaheuristics such as simulated annealing and tabu search. We found that DGA outperformed the competing alternatives in terms of mean error, maximum error, and percentage of least cost solutions. While DGA is computationally intensive, the quality of its solutions is commensurate with the effort expended. In plots of solution quality versus CPU time for the various algorithms evaluated in our study, DGA consistently appeared on the efficient frontier.

Schedule Recovery: Unplanned Absences in Service Operations*

The U.S. service sector loses 2.3% of all scheduled labor hours to unplanned absences, but in some industries, the total cost of unplanned absences approaches 20% of payroll expense. The principal reasons for unscheduled absences (personal illness and family issues) are unlikely to abate anytime soon. Despite this, most labor scheduling systems continue to assume perfect attendance. This oversight masks an important but rarely addressed issue in services management: how to recover from short-notice, short-term reductions in planned capacity. In this article, we model optimal responses to unplanned employee absences in multi-server queueing systems that provide discrete, pay-per-use services for impatient customers. Our goal is to assess the performance of alternate absence recovery strategies under various staffing and scheduling regimes. We accomplish this by first developing optimal labor schedules for hypothetical service environments with unreliable workers. We then simulate unplanned employee absences, apply an absence recovery model, and compute system profits. Our absence recovery model utilizes recovery strategies such as holdover overtime, call-ins, and temporary workers. We find that holdover overtime is an effective absence recovery strategy provided sufficient reserve capacity (maximum allowable work hours minus scheduled hours) exists. Otherwise, less precise and more costly absence recovery methods such as call-ins and temporary help service workers may be needed. We also find that choices for initial staffing and scheduling policies, such as planned overtime and absence anticipation, significantly influence the likelihood of successful absence recovery. To predict the effectiveness of absence recovery policies under alternate staffing/scheduling strategies and operating environments, we propose an index based on initial capacity reserves.

Optimizing Service Attributes: The Seller's Utility Problem*

Service designers predict market share and sales for their new designs by estimating consumer utilities. The services technical features (for example, overnight parcel delivery), its price, and the nature of consumer interactions with the service delivery system influence those utilities. Price and the services technical features are usually quite objective and readily ascertained by the consumer. However, consumer perceptions about their interactions with the service delivery system are usually far more subjective. Furthermore, service designers can only hope to influence those perceptions indirectly through their decisions about nonlinear processes such as employee recruiting, training, and scheduling policies. Like the services technical features, these process choices affect quality perceptions, market share, revenues, costs, and profits. We propose a heuristic for the NP-hard service design problem that integrates realistic service delivery cost models with conjoint analysis. The resulting sellers utility function links expected profits to the intensity of a services influential attributes and also reveals an ideal setting or level for each service attribute. In tests with simulated service design problems, our proposed configurations compare quite favorably with the designs suggested by other normative service design heuristics.

Cross-training performance in flexible labor scheduling environments

Cross-training effectively pools multiple demand streams, improving service levels and, when demand streams are negatively correlated, boosting productivity. When services operate for extended hours, however, those benefits are intermittent because employees take their skills home with them at the end of their shift. This study explores how cross-training and workforce management decisions interact to affect labor costs and service levels in extended hour service operations with uncertain demand and employee attendance. Using a two-stage stochastic model, we first optimally staff, cross-train, schedule, and allocate workers across departments. We then simulate demand and attendance and, as needed, re-allocate available cross-trained workers to best satisfy realized demand. Comparing the performance of full- and partial cross-training policies with that of dedicated specialists, we found that cross-training often, but not always, dominated the performance of a specialized workforce. When cross-trained workers are less proficient than specialists, however, increased cross-training forced tradeoffs between workforce size and capacity shortages. However, both workforce size and service levels often improved with increased scheduling flexibility. Further, increased scheduling flexibility appears to be an efficient strategy for mitigating the effects of absenteeism. Thus, scheduling flexibility may be an important cofactor for exploiting the benefits of cross-training in labor scheduling environments.

Improved network based algorithms for the assembly line balancing problem

In this paper, we present two network based algorithms for solving Type 1 assembly line balancing problems. These algorithms are based on the generation of the network of feasible subsets; the shortest path through this network corresponds to the minimum cost solution. While the methods presented here may require the generation of all feasible subsets, they use upper and lower bounds and dominance to eliminate many of these subsets. The first method (which we call the Frontscan algorithm) evaluates nodes in a manner similar to a procedure originally suggested by Mansoor (1967); the second procedure (which we call the Backscan algorithm) evaluates nodes by proceeding backwards through the network. Both procedures are quite versatile and are easily adapted to the line balancing problem with stochastic task times, duplicate parallel work stations, zoning restrictions, etc. Computational tests indicate that these algorithms are more efficient than previous network based methods (including dynamic programming ...

A dynamic program with fathoming and dynamic upper bounds for the assembly line balancing problem

Abstract It has been suggested that relaxation and fathoming methods can be used to reduce the state space of certain dynamic programs. This paper applies these techniques to the dynamic program for assembly line balancing and shows that with an optimal upper bound, a substantial reduction in state space is possible. With a static nonoptimal upper bound however, the approach is found to offer little improvement over conventional dynamic programming. To achieve more consistent results, a dynamic upper bound procedure is proposed. Applied to a well-known set of assembly line balancing problems, the performance of the proposed algorithm was found comparable to a state-of-the-art integer programming method. The approach appears generalizable to dynamic programs with a similar structure.