Optimum site selection for oil spill response center in the Marmara Sea using the AHP-TOPSIS method

The aim of this study is to determine the optimum location for an oil spill response center in the Marmara Sea. The analytic hierarchy process (AHP) and technique for the order of preference by similarity to ideal solution (TOPSIS) method as the most preferred multi-criteria decision-making (MCDM) technique were used. The results reveal that the LOC criterion and PRA sub-criterion have the highest effects on the optimal location selection of the oil spill response center. According to the results, the most suitable location for the oil spill response center in the region is Izmit Bay Entrance. This location has been proposed for the response headquarters to manage the whole operation with the assistance of auxiliary installations in the area. In this study, only technical and operational variables are taken into account, but political and administrative criteria are excluded. Optimum site selection for oil spill response center in the Marmara Sea using the AHP-TOPSIS method 39 an integer programming model to determine the strategic response operations, which mainly aims to set optimum reaction times. Verma et al. (2013) developed a probabilistic formulation that optimizes the location and capacity of oil spill response centers. Ha (2018) performed a local pollution risk assessment based on accident probability and post-accident sensitivity using the AHP method in Korea. The studies focusing on oil spill matters are mostly concentrated in the strategic regions including straits, narrow channels, gateways and inner seas. For this reason, the Marmara Sea and the Turkish straits have always been in the focus of those studies for years. Guvenet al. (1996) performed a study dealing with the effects of oil pollution from Nassia tanker in Istanbul Strait. Otay and Yenigun (2000) simulated the weathering processes of oil spill from Volganeft-248 in Istanbul Strait. Ors (2003) and Basar (2010) used simulation technique to predict weathering processes of oil in the Marmara Sea and Istanbul Strait. Dogan and Burak (2007) aimed to determine ecological hazard level in Turkish straits and Marmara Sea considering ship originated pollution. Alpar and Unlu (2007) performed environmental risk assessment through “chemical fi ngerprint approach” after the Volganeft-248 accident. Basar et al. (2006) utilized simulation technique to fi nd risky areas for oil spillage after tanker accident at Istanbul Strait. Guven et al. (2007) aimed to determine pollution level of sediments of Turkish Straits and the Marmara Sea using sampling method between 2005 and 2007. Unlu (2007), in his study, characterized the chemical composition of the unknown oil spilled from the Haydarpasa Port through “advanced fi ngerprinting techniques and diagnostic ratio” method. Birpınar et al. (2009) tried to defi ne environmental effects of maritime traffi c on the Istanbul Strait through literature review. Bozkurtoglu (2017) used simulation technique to predict oil spill trajectory in the Istanbul Strait. There are many scientifi c studies in the literature seeking solutions to the facility location problem with multi-criteria decision-making approach models. Yap et al. (2017) and Hong and Xiaohua (2011) utilized the AHP method to set the optimum location of emergency logistics centers. In the solutions of multicriteria decision-making problems, AHP and TOPSIS have been used in many studies integrally. The literature shows that the AHP-TOPSIS method is preferred, especially in terms of solutionsto landfi ll site selection problems (Beskese et al. 2015, Soltanalizadeh et al. 2014, Ertugruland Karakasoglu 2008, Hanine et al. 2016, Yari et al. 2013, Kharat et al. 2016, Hanine et al. 2017). In addition, Gumusay et al. (2016) and Coskun (2016) carried out research aimedatdeterminingthe best location for construction sites of residential areas and marinas.In summary, although there have been many studies in the literature regarding location selection in emergencies, the issue of optimum location of oil spill response centers is a matter of concern. Materials and methods The aim of this study is to determine the optimum location for an oil spill response center in the Marmara Sea usingthe AHP-TOPSIS method. For this purpose, the relevant literature isfi rst reviewed,andthen expert opinions are obtained. The weight of each parameter is calculated using the AHP method. Then, the TOPSIS method is used with the consideration of each alternative’s ranking. Finally, a hypothetical study is performed through a case study to fi nd the optimum location for an oil spill response center in the Marmara Sea. The evaluation process of the study is shown in Fig. 1. Fig. 1. Proposed model of the study Source: Created by authors. 40 B. Koseoglu, M. Buber, A.C. Toz As understood from the fi gure, initially, the decisionmaking problem and study site are identifi ed. After determining the study site, alternative locations for oil spill response centersare identifi ed. Then, the evaluation criteria of the response locations are obtained from the literature review and expert opinions. The AHP method is performed with the data provided and the weight ratios are calculated. Finally, TOPSIS is carried out to calculate the ranking of each alternative. AHP algorithm The purpose of the AHP method is to ensure that the decision-making process is completed in the most effi cient manner, taking into account the individual judgments of the decision-makers and the comparison consistency of the choices in this process, by placing the associated priorities for a given set of options on a scale. This approach supports judgments based on the decision-maker’s knowledge and experience. The AHP provides a simple and effective solution in a multi criteria environment, taking into account all the factors and the systematic way of organizing the countable and uncountable factors (Saaty and Vargas 1982).The steps ofthe AHP method are explained as follows (Saaty 1980); Step 1: In the fi rst stage, the hierarchy explaining the main research problem is established. The hierarchy should include the main goal at the top and alternatives at the bottom. It is important that the number of criteria that impact the endpoint be correctly determined and that detailed descriptions of each criterion are made, so that pairwise comparisons can be made consistently and logically. The standard hierarchic structure of AHP is shown in Fig. 2. Step 2: After the hierarchical model is established, the pairwise comparison decision matrices are constructed to evaluate alternatives on the basis of each factor and determine the importance levels of the factors themselves.