Journal of Quality Engineering and Production Optimization
Volume:1 Issue: 1, Winter - Spring 2015

This study investigates the coordination of production scheduling and maintenance planning in the flow shop scheduling environment. The problem is considered in a bi-objective form, minimizing the makespan as the production scheduling criterion and minimizing the system unavailability as the maintenance planning criterion. The time interval between consecutive maintenance activities as well as the number of maintenance activities on each machine are assumed to be non-fixed. A mixed integer programming formulation of the problem is presented. A special case of the problem, named as single server maintenance is also studied. Then, a bi-objective ant colony system algorithm is presented to solve the problem in focus. To obtain the appropriate components of the proposed algorithm, two sets of experiments are provided. Firstly, experiments are carried out to select the suitable heuristic method to build the heuristic information part of the algorithm between CDS and NEH. Secondly, experiments are reported to select the local search algorithm between iterated local search and adjacent pair-wise interchange. At last, experiments are generated to evaluate the performance of the proposed algorithm, comparing it to the results of an exhaustive search algorithm.

The dynamic facility layout problem (DFLP) is the problem of finding positions of departments on the plant floor for multiple periods (material flows between departments change during the planning horizon) such that departments do not overlap, and the sum of the material handling and rearrangement costs is minimized. In this paper a new optimization algorithm inspired from colonizing weeds, Invasive Weeds Optimization (IWO) is utilized to solve the well-known DFLP. IWO is a simple algorithm which uses basic characteristics of a colony of weeds such as proliferation, growth and competition. A set of reference numerical problems is taken in order to evaluate the efficiency of the algorithm compared with the Dynamic Programming method which had been applied to solve the addressed problem. In order to verify the efficiency of the proposed algorithm a wide range of experiments are carried out to compare the proposed algorithm. Computational results have indicated that the DIWO algorithm is capable of obtaining optimal solutions for small and medium-scaled problems very efficiently.

This paper considers a two-stage assembly flow shop problem (TAFSP) where m machines are in the first stage and an assembly machine is in the second stage. The objective is to minimize a weighted sum of earliness and tardiness time for n available jobs. JIT seeks to identify and eliminate waste components including over production, waiting time, transportation, inventory, movement and defective products.Two-stage assembly flow shop is a combinational production system in which different parts are manufactured on parallel machines independently. This system can be used as a method to produce a variety of products through assembling and combining different set of parts. We apply e-constraint method as an exact approach to validate the proposed model and to obtain fronts of the solutions in the solution spaceThe goal of the proposed problem is trade off between two objectives, minimization makespan and total weighted tardiness and earliness. To analyze effects of n and m factors on the efficiency and performance of the proposed algorithm, we calculate the complexity of sub problems based on factors n and m and the computational results demonstrate that the computational time increases with increasing in n and m, in other words, complexity of the problem increases.

Despite existing various integer programming for sequencing problems, there is not enough information about practical values of the models. This paper considers the problem of minimizing maximum lateness with release dates and presents four different mixed integer programming (MIP) models to solve this problem. These models have been formulated for the classical single machine problem, namely sequenceposition (SP), disjunctive (DJ), linear ordering (LO) and hybrid (HY). The main focus of this research is on studying the structural properties of minimizing maximum lateness in a single machine using MIP formulations. This comparison helps us know the characteristics and priority of different models in minimizing maximum lateness. Regarding to these characteristics and priorities, while solving the lateness problem in the procedure of solving a real-world problem, we apply the lateness model which yields in solution in shortest period of time and try not to use formulations which never lead to solution for large-scale problems. Beside single machine, these characteristics are applicable to more complicated machine environment. We generate a set of test problems in an attempt to solve the formulations, using CPLEX software. According to the computational results, a detailed comparison between proposed MIP formulations is reported and discussed in order to determine the best formulation which is computationally efficient and structurally parsimonious to solve the considering problem. Among the four presented formulations, sequence-position (SP) has the most efficient computational time to find the optimal solution.

In some statistical process control applications, the quality of the product is characterized by the combination of both correlated variable and attributes quality characteristics. In this paper, we propose a novel control scheme based on the combination of two multi-layer perceptron neural networks for simultaneous monitoring of mean vector as well as the covariance matrix in multivariate attribute processes whose quality characteristics are correlated. The proposed neural network-based methodology not only detects separate mean and variance shifts, but also can efficiently detect simultaneous changes in mean vector and covariance matrix of multivariate-attribute processes. The performance of the proposed neural network-based methodology in detecting separate as well as simultaneous changes in the process is evaluated thorough a numerical example based on simulation in terms of average run length criterion and the results are compared with a statistical method based on the combination of two control charts that are developed for monitoring the mean vector and covariance matrix of multivariate-attribute processes, respectively. Theresults of model implementation on numerical example show the superior detection performance of the proposed NN-based methodology rather than the developed combined statistical control charts.

In today’s competitive business environment, the design and management of supply chain network is one of the most important challenges that managers encounter. The supply chain network should be designed for satisfying of customer demands as well as minizing the total system costs. This paper presents a multi-period multi-stage supply chain network design problem under demand uncertainty. The problem is formulated as a two-stage stochastic program. In the first-stage, strategic location decisions are made, while the second-stage contains the tactical decisions. In our developed model, conditional value-at-risk (CVaR) as an effective risk measure is used to produce first-stage decisions in which the loss cost in the second-stage is minimized. In addition, a modified Benders decomposition algorithm is developed to solve the model exactly. The computational results on a set of randomly generated problem instances demonstrate the effectiveness of the proposed algorithm in terms of the solution quality.