S. Sebnem Ahiska

Practical common weight multi-criteria decision-making approach with an improved discriminating power for technology selection

A practical common weight multi-criteria decision-making (MCDM) methodology with an improved discriminating power for technology selection is introduced. The proposed MCDM methodology enables the evaluation of the relative efficiency of decision-making units (DMUs) with respect to multiple outputs and a single exact input. Its robustness and discriminating power are illustrated via a previously reported robot evaluation problem by comparing the ranking obtained by the proposed MCDM framework with that obtained by the cross-efficiency analysis, which is a well-known data envelopment analysis-based methodology. The results indicate that the proposed methodology enables further ranking of data envelopment analysis-efficient DMUs with a notable saving in computations compared with cross-efficiency analysis. Finally, the proposed MCDM framework is extended to incorporate ordinal as well as exact outputs, and an application is presented to illustrate the methodology.

Improved common weight MCDM model for technology selection

This paper presents an improvement on earlier work on a common weight multi-criteria decision-making (MCDM) approach for technology selection by (Karsak, E.E. and Ahiska, S.S., Practical common weight multi-criteria decision-making approach with an improved discriminating power for technology selection. Int. J. Prod. Res., 2005, 43, 1537–1554.) benefiting from a bisection search algorithm. The proposed algorithm enables to calculate the values of discriminating parameter, k, which appears in the introduced efficiency measure, in a systematic and robust manner rather than requiring the decision analyst to assign an arbitrary step size value. In addition, the paper presents comments on the model proposed by (Amin, G.R., Toloo, M. and Sohrabis, B., An improved MCDM DEA model for technology selection. Int. J. Prod. Res., 2006, 44, 2681–2686.) for technology selection. Finally, the robustness of the proposed decision-making framework is illustrated via several numerical examples taken from the above-mentioned ...This paper presents an improvement on earlier work on a common weight multi-criteria decision-making (MCDM) approach for technology selection by (Karsak, E.E. and Ahiska, S.S., Practical common weight multi-criteria decision-making approach with an improved discriminating power for technology selection. Int. J. Prod. Res., 2005, 43, 1537–1554.) benefiting from a bisection search algorithm. The proposed algorithm enables to calculate the values of discriminating parameter, k, which appears in the introduced efficiency measure, in a systematic and robust manner rather than requiring the decision analyst to assign an arbitrary step size value. In addition, the paper presents comments on the model proposed by (Amin, G.R., Toloo, M. and Sohrabis, B., An improved MCDM DEA model for technology selection. Int. J. Prod. Res., 2006, 44, 2681–2686.) for technology selection. Finally, the robustness of the proposed decision-making framework is illustrated via several numerical examples taken from the above-mentioned papers.

Modeling and analysis of a product substitution strategy for a stochastic manufacturing/remanufacturing system

Abstract Many original equipment manufacturers (OEMs) are implementing hybrid manufacturing/remanufacturing systems due to the economic and environmental benefits of remanufacturing such as significant reductions in resource consumption and waste disposal. We consider the periodic-review inventory control problem for such a system where manufactured and remanufactured products are considered non-identical and have separate demand streams. Product returns and demand for both products are stochastic. A remanufactured item has a perceived lower value by the customer and thus has a lower price than a manufactured item. The manufacturer considers the use of a one-way product substitution strategy. When the remanufactured item inventory is out of stock, manufactured items are sold for the remanufactured item price (i.e. lower price) to the customers who demand remanufactured items. The problem is formulated as a discrete-time Markov Decision Process in order to find the optimal inventory policies with substitution and without substitution. The behavior of the system under product substitution strategy and its profitability is investigated through a numerical study based on real data from an automobile parts manufacturer. Results show that profitability is significantly affected by the remanufactured item price (pr) to manufacturing cost (cm) ratio. As pr/cm decreases, the increased profit provided by the substitution strategy over not substituting falls at an increasing rate. However, even when pr/cm ratio goes below 1 (i.e. unit profit by substitution is negative), substitution may still be profitable due to savings in lost sales.

A common-weight MCDM framework for decision problems with multiple inputs and outputs

This paper presents a common weight multi-criteria decision making (MCDM) approach for determining the best decision making unit (DMU) taking into consideration multiple inputs and outputs. Its robustness and discriminating power are illustrated through comparing the results with those obtained by data envelopment analysis (DEA) and its extensions such as cross efficiency analysis and minimax efficiency DEA model, which yield a ranking with an improved discriminating power. Several examples reported in earlier research addressing DEAs discriminating power are used to illustrate the application of the proposed approach. The results indicate that the proposed framework enables further ranking of DEA-efficient DMUs with a notable saving in the number of mathematical programming models solved.

Heuristic inventory policies for a hybrid manufacturing/remanufacturing system with product substitution

Abstract We consider inventory control for an infinite-horizon hybrid manufacturing/remanufacturing system with product substitution under stochastic demand and returns. Remanufactured and manufactured products are considered as different products having different costs and selling prices as well as separate demand streams. Remanufactured products have a higher stock out risk than manufactured products because the remanufacturing capacity is dependent on available returns for remanufacture. One way to cope with this stock-out risk is to use a downward substitution strategy, under which a manufactured product (i.e. higher value item) substitutes for a remanufactured product (i.e. lower value item) in case the latter runs out of stock. This problem can be formulated as a Markov Decision Process in order to determine the optimal manufacturing and remanufacturing decisions under product substitution. However, the optimal policy has a complicated structure. Based on characterization of the optimal policies, we propose intuitive heuristic policies that are easy to implement in practice. Then, we develop a heuristic search technique to determine the parameter values for these policies in an efficient way. We evaluate the performance of the proposed heuristic policies compared to the optimal inventory policy through a real case study involving a spare part manufacturer.

Analysis of Downward Product Substitution in a Recoverable System

We consider the inventory control problem for an infinite-horizon stochastic hybrid manufacturing /remanufacturing system with product substitution under stochastic demand and returns. Remanufactured and manufactured products are considered as two different products, having different costs and selling prices as well as separate demand streams. Remanufactured products have a higher stock out risk because the remanufacturing capacity is mainly dependent on the amount of returns available for remanufacture. One way to cope with the stock-out issue for remanufactured products is to use a downward substitution strategy, which allows a manufactured product (i.e. higher value item) to be substituted for a remanufactured product (i.e. lower value item) in case the latter runs out of stock. We formulate this problem as Markov Decision Process in order to determine the optimal manufacturing and remanufacturing decisions under product substitution, and through numerical experimentation, we investigate the effects of stochastic demand/return distributions on the profitability of the substitution strategy.

Analysis of a Downward Substitution Strategy in a Manufacturing/Remanufacturing System

A hybrid production system is considered where both manufacture of new product and remanufacture of returned items is performed. Due to consumer perception, new and remanufactured products are treated as different products with different costs and selling prices as well as separate demand streams. Remanufactured products have a higher stock out risk because the remanufacturing capacity is limited by the amount of returns available for remanufacture. One way to cope with this risk is to use a downward substitution strategy, i.e. a higher valued manufactured product is substituted for an out of stock lower valued remanufactured product. We formulate this control problem as an infinite-horizon hybrid manufacturing/remanufacturing system with product substitution under stochastic demand and returns. We model it as a Markov Decision Process in order to determine the optimal manufacturing and remanufacturing decisions under product substitution. The effects of stochastic demand/return distributions on the profitability of the substitution strategy are investigated through numerical experimentation.