International Journal of Industrial Engineering and Productional Research
Volume:20 Issue: 1, Jun 2009

The problem of lot sizing، sequencing and scheduling multiple products in flow line production systems has been studied by several authors. Almost all of the researches in this area assumed that setup times and costs are sequence –independent even though sequence dependent setups are common in practice. In this paper we present a new mixed integer non linear program (MINLP) and a heuristic method to solve the problem in sequence dependent case. Furthermore، a genetic algorithm has been developed which applies this constructive heuristic to generate initial population. These two proposed solution methods are compared on randomly generated problems. Computational results show a clear superiority of our proposed GA for majority of the test problems.

Fuzzy Cognitive Maps (FCMs) have successfully been applied in numerous domains to show the relations between essential components in complex systems. In this paper, a novel learning method is proposed to construct FCMs based on historical data and by using meta-heuristic: Genetic Algorithm (GA), Simulated Annealing (SA), and Tabu Search (TS). Implementation of the proposed method has demonstrated via real data of a purchase system in order to simulate the system’s behavior.

Modern business organizations have appreciated the significance of having competitive advantage through the delivery of continuous improvement towards the customers, and being knowledge-oriented. Indisputably, Knowledge Management (KM) plays a key role in the success of Customer Relationship Management (CRM). In this regard, Customer Knowledge Management (CKM) is a newly developed concept that deals with knowledge from customers rather than knowledge about customers. However, little research has been done on the application of CKM in e-business. In this paper, after an overview of the literature, an application of CKM in Customer Lifetime Value (CLV) measurement is studied in an e-retailer case where Corporate Image and Reputation are taken into consideration.

Selecting an effective project plan is a significant area in the project management. The present paper introduces a technique to identify the project plan efficient frontier for assessing the alternative project plans and selecting the best plan. The efficient frontier includes two criteria: the project cost and the project time. Besides, the paper presents a scheme to incorporate Directed Acyclic Graph (DAG) into the project risk analysis. This scheme is used to estimate the expected impacts of the occurrence of the project risks on the project cost and the project time. Also, a theoretical model is defined to provide integration between project risk analysis and overall project planning using the breakdown structures. We believe that applying the proposed technique helps the company’s managers in most effective manner dealing with his complicated project plan assessment and selection problems. The application of the technique was implemented in the companies in construction industry in which represented a considerable cost and time improvements.

Multiple Objective Programming (MOP) problems have become famous among many researchers due to more practical and realistic implementations. There have been a lot of methods proposed especially during the past four decades. In this paper, we develop a new algorithm based on a new approach to solve MOP problems by starting from a utopian point (which is usually infeasible) and moving towards the feasible region via stepwise movements and a plain continuous interaction with Decision Maker (DM). We consider the case where all objective functions and constraints are linear. The implementation of the proposed algorithm is demonstrated with two numerical examples.

In this study, an intelligent decision model based on reliability measures is proposed for the traction control system of the speed sensor vehicles. This model is formulated for the integrated and complex system based on Bayesian Decision Theory that is used to model pattern recognition of train traction conditions along with high precision and manageable number of; state natures (i.e., spin/slip, normal, and slide), features (delta speed and train speed) and having the prior knowledge traction conditions. The intelligent model is useful to alleviate the impact of the noisy sensors (inaccurate data) and its delays in such a hard real-time control system. The model’s engine involves mathematical problem which can be solved in any programming language in on-board or embedded computers. The conceptual model is applied to a hypothetical case study with promising results for target and simulation systems.

In this paper we focus on a continuously deteriorating two units series equipment which its failure can not be measured by cost criterion. For these types of systems avoiding failure during the actual operation of the system is extremely important. In this paper we determine inspection periods and maintenance policy in such a way that failure probability is limited to a pre-specified value and then optimum policy and inspection period are obtained to minimize long-run cost per time unit. The inspection periods and maintenance policy are found in two phases. Failure probability is limited to a pre-specified value In the first phase, and in the second phase optimum maintenance thresholds and inspection periods are obtained in such a way that minimize long-run expected.

Selection of an appropriate set of Information System (IS) projects is a critical business activity which is very helpful to all organizations. In this paper, after describing real IS project selection problem of Iran Ministry of Commerce (MOC), we introduce two Data Envelopment Analysis (DEA) models. Then, we show applicability of introduced models for identifying most efficient IS project from 8 competing projects. Then, in order to provide further insight, results of two introduced models are compared. It is notable that using basic DEA models -CCR and BCC- decision maker is not able to find most efficient Decision Making Unit (DMU) since these models identify some of DMUs as efficient which their efficiency scores equal to 1. As an advantage, the applied models can identify most efficient IS (in constant and variable return to scale situations) by solving only one linear programming (LP). So these models are computationally efficient. It is while using the basic DEA models requires decision maker to solve a LP for each IS.