SCIENTIA IRANICA
Year:2010 | Volume:17 | Issue:1 (TRANSACTION E: INDUSTRIAL ENGINEERING)|Papers Count:7

An electromagnetism algorithm is a meta-heuristic proposed to derive approximate solutions for computationally hard problems. In the literature, several successful applications have been reported for graph-based optimization problems, such as scheduling problems. This paper presents an application of the electromagnetism algorithm to supplier selection in a production planning process where there are multiple products and multiple customers and also capacity constraints. We consider a situation where the demand quantity of multiple discrete products is known over a planning horizon. The required raw material for each of these products can be purchased from a set of approved suppliers. Also, a demand-dependent delivery time (due date) and maximum delivery time (deadline) apply for each demand. Problems containing all these assumptions have not been addressed previously in the literature. A decision needs to be made regarding what raw material to order and in what quantities, which suppliers and, finally, at which periods. Numerical results indicate that the electromagnetism algorithm exhibits impressive performances with small error ratios. The results support the success of the electromagnetism algorithm application to the supplier selection problem of interest.

In many practical situations, the quality of a process or product can be characterized by a function or profile. Here, we consider a polynomial profile and investigate the effect of the violation of a common independence assumption, implicitly considered in most control charting applications, on the performance of the existing monitoring techniques. We specifically consider a case when there is autocorrelation between profiles over time. An autoregressive model of order one is used to model the autocorrelation structure between error terms in successive profiles. In addition, two remedial methods, based on time series approaches, are presented for monitoring autocorrelated polynomial profiles in phase II. Their performances are compared using a numerical simulation runs in terms of an Average Run Length (ARL) criterion. The effects of assignable cause and autocorrelation coefficient on the shape of aprofiles are also investigated.

This paper deals with the resource-constrained project scheduling problem with flexible work profiles. In this problem, a project contains activities interrelated by finish-start-type precedence constraints with a time lag of zero. In many real-life projects, however, it often occurs that only one renewable bottleneck resource is available and that activities do not have a fixed prespecified duration and associated resource requirement, but a total work content, which essentially indicates how much work has to be performed on them. Based on this work content, all feasible work profiles have to be specified for the activities, each characterized by a fixed duration and a resource requirement profile. The task of the problem is to find the optimum work profile and start time of each activity in order to minimize the project makespan. We propose a procedure to find all feasible work profiles of each activity and we use a genetic algorithm with a new crossover operator to schedule the activities. Computational results on a randomly generated problem set are presented.

The location-routing problem is one of the most important location problems for designing integrated logistics systems. In the last three decades, various types of objective function and constraints have been considered for this problem. However, time window constraints have received little attention, despite their numerous real-life applications. In this article, a new 4-index mathematical model, an efficient and fast heuristic and a lower bound for the two-echelon location-routing problems with soft time window constraints are presented. The proposed heuristic tries to solve the problem via creating an initial solution, then improving it by searching on six neighborhoods of the solution, and using the Or-opt heuristic. At the end, computational results show the efficiency of the proposed heuristic, using the proposed lower bound.

The Shewhart np control chart is often used to monitor the quantity of nonconforming, but it is slow in detecting small deviations. This paper proposes an efficient approach to monitor the quantity of nonconforming. The novelty of the paper is utilization of an initial belief to construct an analytic variable limit np control chart. The proposed method uses all gathered data, sequentially. This approach is significantly faster than some existent effective approaches in detecting small deviations. These charts are mainly used for evaluation of the initial setup in the process. The simulated results for the average run  length profiles demonstrate the superiority of the new approach against the standard np chart, binomial CUSUM, binomial EWMA and moving average approach.

In this paper, an Economic Production Quantity (EPQ) model is studied, in which the production defective-rate follows either a uniform or a normal probability distribution. Shortages are allowed and take a backorder state, and the existence of only one machine causes a limited production capacity for the common cycle length of all products. The aim of this research is to determine the optimal production quantity of each product, such that the expected total cost including holding, shortage, production, setup and defective items cost is minimized. The mathematical model of the problem is derived, for which the objective function is proved to be convex. Then, a derivative approach is utilized to obtain the optimal solution. At the end, two numerical examples are provided to illustrate the practical usage of the proposed method.

A holding or a multi-business corporate seeks to coordinate its supply for minimum overall costs. A Corporate Supply Optimizer (CSO), as a central entity taking advantage of the notion of flow networks, gathers necessary operational information from members of the corporate supply chain. The CSO then guides supply chain members on ordering decisions for a minimum overall cost for the entire supply chain. Its computational engine models the entire supply chain with multiple members in four stages to satisfy customer demand. The CSO seeks a solution with minimum total costs, unlike non cooperative supply chains where individual members compete to optimize their local costs. The existing literature stays with restrictive assumptions on the number of supply chain stages, disallowing a case of multiple products. Simulation results indicate an approximately 26% reduction in total costs of the supply chain utilizing the CSO.