Heuristics with guaranteed performance bounds for a manufacturing system with product recovery

We consider a manufacturing system with product recovery. The system manufactures a new product as well as remanufactures the product from old, returned items. The items remanufactured with the returned products are as good as new and satisfy the same demand as the new item. The demand rate for the new item and the return rate for the old item are deterministic and constant. The relevant costs are the holding costs for the new item and the returned item, and the fixed setup costs for both manufacturing and remanufacturing. The objective is to determine the lot sizes and production schedule for manufacturing and remanufacturing so as to minimize the long-run average cost per unit time. We first develop a lower bound among all classes of policies for the problem. We then show that the optimal integer ratio policy for the problem obtains a solution whose cost is at most 1.5% more than the lower bound.     

Filtering Policies in Loss Queuing Network Location Problems

In this paper we deal with two stationary loss queuing network location models. We analyze the influence of filtering policies on the locational aspect of the problems. We assume that requests for service are placed at nodes of a transportation network and they arrive in time as independent homogeneous Poisson processes with different input rates. The considered policies only cover a given proportion of requests even if there are idle service units. This proportion is stationary and fixed in advance and only depends on the node where the request is originated. The objective is to find the location of the facilities together with the filtering policy to be applied that minimize the expected total cost per unit time with respect to a given cost structure. Properties and computational results are presented enabling the resolution of these problems efficiently and showing the good performance of filtering policies in terms of both the overall operating costs, and the demand that is served.     

Dynamic admission and service rate control of a queue

This paper investigates a queueing system in which the controller can perform admission and service rate control. In particular, we examine a single-server queueing system with Poisson arrivals and exponentially distributed services with adjustable rates. At each decision epoch the controller may adjust the service rate. Also, the controller can reject incoming customers as they arrive. The objective is to minimize long-run average costs which include: a holding cost, which is a non-decreasing function of the number of jobs in the system; a service rate cost c(x), representing the cost per unit time for servicing jobs at rate x; and a rejection cost κ for rejecting a single job. From basic principles, we derive a simple, efficient algorithm for computing the optimal policy. Our algorithm also provides an easily computable bound on the optimality gap at every step. Finally, we demonstrate that, in the class of stationary policies, deterministic stationary policies are optimal for this problem.     

3.1. OR in banking

In this paper, we describe a deterministic multiperiod capacity expansion model in which a single facility serves the demand for many products. Potential applications for the model can be found in the capacity expansion planning of communication systems as well as in the production planning of heavy process industries. The model assumes that each capacity unit simultaneously serves a prespecified (though not necessarily integer) number of demand units of each product. Costs considered include capacity expansion costs, idle capacity holding costs, and capacity shortage costs. All cost functions are assumed to be nondecreasing and concave. Given the demand for each product over the planning horizon, the objective is to find the capacity expansion policy that minimizes the total cost incurred. We develop a dynamic programming algorithm that finds optimal policies. The required computational effort is a polynomial function of the number of products and the number of time periods. When the number of products equals one, the algorithm reduces to the well-known algorithm for the classical dynamic lot size problem.     

Using aggregate fill rate for dynamic scheduling of multi-class systems

For dynamic scheduling of multi-class systems where backorder cost is incurred per unit backordered regardless of the time needed to satisfy backordered demand, the following models are considered: the cost model to minimize the sum of expected average inventory holding and backorder costs and the service model to minimize expected average inventory holding cost under an aggregate fill rate constraint. Use of aggregate fill rate constraint in the service model instead of an individual fill rate constraint for each class is justified by deriving equivalence relations between the considered cost and service models. Based on the numerical investigation that the optimal policy for the cost model is a base-stock policy with switching curves and fixed base-stock levels, an alternative service model is considered over the class of base-stock controlled dynamic scheduling policies to minimize the total inventory (base-stock) investment under an aggregate fill rate constraint. The policy that solves this alternative model is proposed as an approximation of the optimal policy of the original cost and the equivalent service models. Very accurate heuristics are devised to approximate the proposed policy for given base-stock levels. Comparison with base-stock controlled First Come First Served (FCFS) and Longest Queue (LQ) policies and an extension of LQ policy (Δ policy) shows that the proposed policy performs much better to solve the service models under consideration, especially when the traffic intensity is high.     

Polynomial cases of the economic lot sizing problem with cost discounts

In this paper we study the economic lot sizing problem with cost discounts. In the economic lot sizing problem a facility faces known demands over a discrete finite horizon. At each period, the ordering cost function and the holding cost function are given and they can be different from period to period. There are no constraints on the quantity ordered in each period and backlogging is not allowed. The objective is to decide when and how much to order so as to minimize the total ordering and holding costs over the finite horizon without any shortages. We study two different cost discount functions. The modified all-unit discount cost function alternates increasing and flat sections, starting with a flat section that indicates a minimum charge for small quantities. While in general the economic lot sizing problem with modified all-unit discount cost function is known to be NP-hard, we assume that the cost functions do not vary from period to period and identify a polynomial case. Then we study the incremental discount cost function which is an increasing piecewise linear function with no flat sections. The efficiency of the solution algorithms follows from properties of the optimal solution. We computationally test the polynomial algorithms against the use of CPLEX.     

Simple heuristics for push and pull remanufacturing policies

Inventory policies for joint remanufacturing and manufacturing have recently received much attention. Most efforts, though, were related to (optimal) policy structures and numerical optimization, rather than closed form expressions for calculating near optimal policy parameters. The focus of this paper is on the latter. We analyze an inventory system with unit product returns and demands where remanufacturing is the cheaper alternative for manufacturing. Manufacturing is also needed, however, since there are less returns than demands. The cost structure consists of setup costs, holding costs, and backorder costs. Manufacturing and remanufacturing orders have non-zero lead times. To control the system we use certain extensions of the familiar (s, Q) policy, called push and pull remanufacturing policies. For all policies we present simple, closed form formulae for approximating the optimal policy parameters under a cost minimization objective. In an extensive numerical study we show that the proposed formulae lead to near-optimal policy parameters.     

The role of repair strategy in warranty cost minimization: An investigation via quasi-renewal processes

Most companies seek efficient rectification strategies to keep their warranty related costs under control. This study develops and investigates different repair strategies for one- and two-dimensional warranties with the objective of minimizing manufacturer’s expected warranty cost. Static, improved and dynamic repair strategies are proposed and analyzed under different warranty structures. Numerical experimentation with representative cost functions indicates that performance of the policies depend on various factors such as product reliability, structure of the cost function and type of the warranty contract.     

Newsvendor equations for production networks

We consider production networks with stochastic activity leadtimes. When activities finish early holding costs are incurred and when end products are delivered late penalty costs are incurred. Objective is to find the activity start and finish times that minimize the total cost. We introduce the concept of a tardy path and derive the optimality equations for each node in the network. We show that under the optimal solution, for a set of nodes the tardiness probability satisfies the Newsvendor equations.     

Joint optimization of process improvement investments for supplier–buyer cooperative commerce

This research focuses on supporting the formation of strategic alliances through the concept of cooperative commerce, where suppliers and buyers work together to jointly optimize their business. The general goal of this research is to examine existing cooperative commerce models for obstacles that would hinder their successful implementation into modern industrial applications and to address those shortcomings. Total annual cost equations are formulated to capture the joint total relevant cost of cooperative commerce business relationships. These total joint relevant cost models will include terms that capture the ordering cost, holding cost, and cost of quality, as well as any applicable investment cost for process improvements, consistent with traditional economic order quantity and economic production quantity theory. This research corrects a modelling/computational error found in the literature that led to underestimation of the effectiveness of process improvements in joint economic lot size models by 5–12%. In addition, the models are expanded to accommodate a full range of product quality inspection policies, from 0 to 100% product inspections. Furthermore, the models are modified to account for the cost of scrap generation, as well as the effects of accepting non-conforming product and rejecting conforming product during quality inspections. Once the total cost models are expanded to account for these neglected costs, the joint total relevant cost equations are minimized to find the optimal batch sizes, and the effects of each model extension on the model solution are studied. Results indicate that these extensions have a significant impact on the model results, such as reduced optimal batch sizes and increased optimal fraction conforming product.     

Marginal productivity index policies for scheduling a multiclass delay-/loss-sensitive queue

This paper addresses the problem of scheduling a Markovian multiclass queue with a finite dedicated buffer for each class, where class-dependent linear holding and rejection cost rates model differing levels of tolerance to delay and loss. The goal is to design well-grounded and tractable scheduling policies that nearly minimize expected total discounted or long-run average cost. New dynamic index policies are introduced, awarding higher priority to classes with larger index values, where a class’ index measures the marginal productivity of work at its current state. The results are obtained by deploying the work-cost analysis approach to marginal productivity indices (MPIs) for restless bandits developed by the author, which is extended to the bias criterion. The MPI furnishes new insights: for a loss-sensitive class, it is a decreasing function of the number of empty buffer spaces, independent of the buffer size; for a delay-sensitive class, it is a decreasing function of the queue length. Such opposite orderings show that preventive work is more valuable than reactive work for the latter classes, whereas the opposite holds for the former. The results of a computational study on two-class instances are reported, shedding light on how the MPI policy’s relative performance varies with each parameter. Parameter ranges are thus identified where the MPI policy is near optimal, and substantially outperforms conventional benchmark policies. 2000 Mathematics Subject Classification 90B22 · 90B36 · 90B18 · 60K25 · 60K30 · 68M20     

Stochastic Multiproduct Inventory Models with Limited Storage

This paper studies multiproduct inventory models with stochastic demands and a warehousing constraint. Finite horizon as well as stationary and nonstationary discounted-cost infinite-horizon problems are addressed. Existence of optimal feedback policies is established under fairly general assumptions. Furthermore, the structure of the optimal policies is analyzed when the ordering cost is linear and the inventory/backlog cost is convex. The optimal policies generalize the base-stock policies in the single-product case. Finally, in the stationary infinite-horizon case, a myopic policy is proved to be optimal if the product demands are independent and the cost functions are separable.     

Optimality conditions for regenerative inventory systems under batch demands

We consider a single-item, continuous-review, (s, S) inventory system, under complete backlogging and a constant procurement lead time. Demands occur in independent, identically distributed batches, separated by independent identically distributed intervals. The model also includes a class of periodic review systems as a special case. Of interest are the optimal control policies with respect to a stationary cost rate function, constructed to include ordering, holding and shortage costs. We study the structural properties of the cost rate function and report some new bounds and optimality conditions. An application of these to the computation and approximation of optimal policies is also discussed.     

Optimal production inventory policy for defective items with fuzzy time period

In this paper, an optimal production inventory model with fuzzy time period and fuzzy inventory costs for defective items is formulated and solved under fuzzy space constraint. Here, the rate of production is assumed to be a function of time and considered as a control variable. Also the demand is linearly stock dependent. The defective rate is taken as random, the inventory holding cost and production cost are imprecise. The fuzzy parameters are converted to crisp ones using credibility measure theory. The different items have the different imprecise time periods and the minimization of cost for each item leads to a multi-objective optimization problem. The model is under the single management house and desired inventory level and product cost for each item are prescribed. The multi-objective problem is reduced to a single objective problem using Global Criteria Method (GCM) and solved with the help of Fuzzy Riemann Integral (FRI) method, Kuhn–Tucker condition and Generalised Reduced Gradient (GRG) technique. In optimum results including production functions and corresponding optimum costs for the different models are obtained and then are presented in tabular forms.     

Integrating transportation and production: an international study case

The problem we study is inspired by the real case of Mesdan S.p.A., an Italian company worldwide leader in the textile machinery sector, which has two production units with two warehouses, one located in Italy (Brescia) and the other in China (Foshan). The critical point in this logistic system is the integration between production and transportation management, given the long distance between Brescia and Foshan. Shipments are performed by the means of different types of vehicles with different unit costs and significantly different lead times. Variable production costs, variable and fixed transportation costs and, possibly, inventory costs are charged in the objective function. Different production policies are compared. Our aim is to determine integrated policies that minimize the total cost of the system. We formulate integer linear programming models for the solution of these problems, and we solve the real instance and carry out a sensitivity analysis of the optimal solution.     

Designing a new computational approach of partial backlogging on the economic production quantity model for deteriorating items with non-linear holding cost under inflationary conditions

This paper discusses the production inventory model over an infinite time horizon. Here we consider demand as a function of stock and time. Deterioration is a function of time and time-varying production. Our objective is to minimize the total cost which is a function of set up cost, holding cost, shortage cost, and opportunity cost due to lost sales. The traditional costs such as purchasing cost, shortage cost and opportunity cost due to lost sales are kept constant. We consider holding cost to be a non-linear function of time. Shortages are allowed and are partially backlogged. Here, time durations are the decision variables. Numerical examples are given to illustrate the model.     

Optimal control of a divergent multi-echelon inventory system

Consider a divergent multi-echelon inventory system, such as a distribution system or a production system. At every facility in the system orders are placed (or production is initiated) periodically. The order arrives after a fixed lead time. At the end of each period linear costs are incurred penalty costs are incurred at the most downstream facilities for back-orders. The objective is to minimize the expected holding and penalty costs per period. We prove that under the balance assumption it is cost optimal to control every facility by an order-up-to-policy. The optimal replenishment policy, i.e. the order-up-to-level and the allocation functions at each facility, can be determined by system decomposition. This decomposition reduces complex multi-dimensional control problems to simple one-dimensional problems.     

On the optimality of a full-service policy for a queueing system with discounted costs

We provide weak sufficient conditions for a full-service policy to be optimal in a queueing control problem in which the service rate is a dynamic decision variable. In our model there are service costs and holding costs and the objective is to minimize the expected total discounted cost over an infinite horizon. We begin with a semi-Markov decision model for a single-server queue with exponentially distributed inter-arrival and service times. Then we present a general model with weak probabilistic assumptions and demonstrate that the full-service policy minimizes both finite-horizon and infinite-horizon total discounted cost on each sample path.     

Analysis of job transfer policies in systems with unreliable servers

We consider a system where incoming jobs may be executed at different servers, each of which goes through alternating periods of being available and unavailable. Neither the states of the servers nor the relevant queue sizes are known at moments of arrival. Hence, a load balancing mechanism that relies on random time-out intervals and job transfers from one queue to another is adopted. The object is to minimize a cost function which may include holding costs and transfer costs. A model of a single queue with an unreliable server and timeouts is analyzed first. The results are then used to obtain an approximate solution for arbitrary number of queues. Several transfer policies are evaluated and compared.     

A Simple and Effective Component Procurement Policy for Stochastic Assembly Systems

We examine the component procurement problem in a single-item, make-to-stock assembly system. The suppliers are uncapacitated and have independent but non-identically distributed stochastic delivery lead times. Assembly is instantaneous, product demand follows a Poisson process and unsatisfied demand is backordered. The objective is to minimize the sum of steady-state holding and backorder costs over a pre-specified class of replenishment policies. To keep the analysis tractable, we impose a synchronization assumption that no mixing occurs between sets of component orders. Combining existing results from queueing theory with original results concerning distributions that are closed under maximization and translation, we derive a simple approximate solution to the problem when lead time variances are identical. In simulations, our derived policy is within 2% of optimal and significantly outperforms policies that ignore either component dependence or lead time stochasticity. It is also quite robust with respect to various model assumptions, except the synchronization one.