Roghaye Hemmatjou, Nasim Nahavandi, Behzad Moshiri,
Volume 27, Issue 3 (IJIEPR 2016)
Abstract
In most of the multi–criteria decision–analysis (MCDA) problems in which the Choquet integral is used as aggregation function, the coefficients of Choquet integral (capacity) are not known in advance. Actually, they could be calculated by capacity definition methods. In these methods, the preference information of decision maker (DM) is used to constitute a possible solution space. The methods which are based on optimizing an objective function most often suffer from three drawbacks. Firstly, the selection of the ultimate solution from solution set is arbitrarily done. Secondly, the solution may provide more information than whatever proposed by DM. Thirdly, DM may not fully interpret the results. Robust capacity definition methods are proposed to overcome these kinds of drawbacks, on the other hand these methods do not consider evenness (uniformity) which is a major property of capacity. Since in capacity definition methods, the preference information on only a subset of alternatives called reference alternatives, is used, defining the capacity as uniform as possible could improve its capability in evaluating non–reference alternatives. This paper proposes an algorithm to define a capacity that is based only on the preference information of DM and consequently is representative. Furthermore, it improves evenness of capacity and consequently its reliability in evaluating non–reference alternatives. The algorithm is used to evaluate power plant projects. Power plant projects are of the most important national projects in Iran and a major portion of national capital is invested on them, so these projects should be scientifically evaluated in order to figure out their performance. Case–specific criteria are considered in addition to general criteria used in project performance evaluation. The evaluation results obtained from proposed algorithm are compared with those of the most representative utility function method.
Adeleh Behzad, Mohammadali Pirayesh, Mohammad Ranjbar,
Volume 28, Issue 3 (IJIEPR 2017)
Abstract
In last decades, mobile factories have been used due to their high production capability, carrying their equipment and covering rough and uneven routes. Nowadays, more companies use mobile factories with the aim of reducing the transportation and manufacturing costs. The mobile factory must travel between the suppliers, visit all of them in each time period and return to the initial location of the mobile factory. In this paper, we present an integer nonlinear programming model for production scheduling and routing of mobile factory with the aim of maximization of profit. This problem is similar to the well-known Traveling Salesman Problem (TSP) which is an NP-hard problem. Also at each supplier, the scheduling problem for production is NP-hard. After linearization, we proposed a heuristic greedy algorithm. The efficiency of this heuristic algorithm is analyzed using the computational studies on 540 randomly generated test instances. Finally, the sensitivity analysis of the production cost, transportation cost and relocation cost was conducted.
Shadan Sadighbehzadi, Zohreh Moghaddas, Amirreza Keyghobadi, Mohsen Vaez-Ghasemi,
Volume 29, Issue 4 (IJIEPR 2018)
Abstract
Natural disasters and crisis are inevitable and each year impose destructive effects on human as injuries and damage to property. In natural disasters and after the outbreak of the crisis, demand for logistical goods and services increase. Effective distribution of emergency aid could have a significant role in minimizing the damage and fatal accident. In this study, a three-level relief chain including a number of suppliers in fixed locations, candidate distribution centers and affected areas at certain points are considered. For this purpose a mixed integer nonlinear programming model is proposed for open transportation location routing problem by considering split delivery of demand. In order to solve a realistic problem, foregoing parameters are considered as fuzzy in our proposed mode. The objectives of the proposed model include total cost minimization, minimization of the maximum travel time of
vehicles and minimization of unmet demands. In order to solve the problem of the proposed model, fuzzy multi-objective planning is used. For efficiency and effectiveness of the proposed model and solution approach, several numerical examples are studied. Computational results show the effectiveness and efficiency of the model and the proposed approach.
