Comparison of multi-criteria-decision-making tools for supplier selection in the construction industry

Suppliers are the providers of the materials in all projects and they have a strategic role on achieving project aims and protecting competitiveness levels of companies in every sector. Companies' main expectations from their suppliers are generally same in all sectors, and the expectations from them are mainly receiving products with expected quality, on the right time and within budget. The absence of the materials or receiving wrong amount of materials which are necessary for production, may lead delays and cost increasing on the projects. Therefore, selecting the right supplier for the right job is crucial.Supplier selection process in the construction industry shows similarities with other sectors. Success in the construction projects can be roughly describe as finishing the work on time, within budget and with the expected quality. Therefore, working with the right supplier who is the responsible for delivering the materials, can highly affect the success of the projects. In order to finishing the construction work successfully, construction companies should select their suppliers correctly. Due to suppliers have great effects on projects success, supplier selection process should be based on mathematical analysis rather than project managers's instincts and experience. Supplier selection problems can be defined as a multi-critera decision making problem because decision makers of the companies often evaluate their suppliers according to several criteria. The aim of this study is proposing and comparing several multi-criteria decision making tools to help construction companies on selecting the proper supplier for their projects. For this purpose an extensive literature review was conducted and explained in the first and second section of this thesis. In the literature several studies dealing with supplier selection problems was found. These studies can be categorized into two main groups; the studies that aimed to find out both the criteria that are important in the supplier selection problems and their importance levels, and the studies that proposed several tools, methods and methodologies for supplier selection problems. Each of these previous studies have improved the supplier selection process so far. After sharing the findings of the literature review, seven multi-criteria decision making techniques that can be used in both supplier selection problems and similar decision making problems, was explained in the third and fourth section of this thesis. Both fuzzy techniques and techniques that works with crisps numbers were selected and explained step by step. Four of the selected techniques, namely AHP, ELECTRE III, PROMETHEE and TOPSİS, are working with crisp numbers. The other three techniques, which are: Fuzzy AHP, Fuzzy ELECTRE and Fuzzy TOPSİS are fuzzy techniques that are working with fuzzy numbers. Fuzzy numbers were explained in the beginning of the fuzzy decision making techniques section. In the fifth section, the selected techniques were applied to a problem of selection the rail supplier for an intercity railway project in Saudi Arabia. First an interview with one of the tendering engineer in the railway company were done to determine the selection criteria that were used in rail supplier selection process. After the interview, eight selection criteria namely supplier' product quality, supplier' lead time, past and current relationships with supplier, the total price of the product, payment facilities offered by the supplier, communication with supplier, production capacity of supplier and technical capability of the supplier, were determined. Also, five alternative rail suppliers for this project were selected. In the light of this information gained at the interview, the problem hierarchy was created.After identifying both the selection criteria and alternative suppliers, a questionnaire, which comprises four section, was designed. The first section includes one question matrix that aims to evaluate suppliers by each criteria. In the second section, real information of alternative suppliers about their total price, lead time and production capacity were asked. Third section has one question that aim to investigate the threshold values of the selection criteria. In the fourth section 9 pairwise comparison matrix were asked. The questionnaire was conducted among four tendering engineers who were decision makers of supplier selection process for this railway project. The data gathering from the questionnaire was used in solving rail supplier selection problem.First analytical hierarchy process (AHP) was used to calculate the importance weights of the selection criteria. Then, the ranking of the suppliers were determined by the same method. Only the data coming from the pairwise comparison matrixes were used in this method. Superdecisions software were used for calculations of AHP method. After that, ELECTRE III were used for calculating the ranking of the suppliers. In ELECTRE III method, the weights of the selection criteria and the data coming from the questionnaire's first three section were used. The real veto threshold values of the selection criteria were also used in this technique. Veto threshold values were not used in any other techniques. Lamsade program were used in ELECTRE III calculations. Later, PROMETHEE and TOPSİS methods were used to rank the alternatives. The data that were used in ELECTRE III method were also used in these methods with the exception of threshold values. In PROMETHEE method only preference and indifference thresholds were used while no threshold values were used in TOPSIS method. The Visual Promethee program were used in solutions of PROMETHEE method while excel tables were created for TOPSİS method. After calculating the ranking of the alternative suppliers by using crisp numbers, fuzzy methods were used, too. To be able to use the values coming from the questionnaire in the fuzzy methods, fuzzification procedure which is the way of converting crisp numbers to fuzzy numbers, were done. Fuzzy AHP, Fuzzy ELECTRE and Fuzzy TOPSIS methods were used to calculate the ranking of the suppliers after converting the crisp data to fuzzy numbers. Only fuzzy pairwise comparison matrixes data were used in Fuzzy AHP method and both fuzzy critera weight values and ranking of the suppliers were calculated. In both Fuzzy ELECTRE and Fuzzy Topsis methods, the fuzzy criteria weights that were calculated by Fuzzy AHP method were used. Real threshold values were not used in Fuzzy ELECTRE method. No threshold values were used in Fuzzy TOPSIS method as same as TOPSIS method.There were no software for Fuzzy AHP, Fuzzy ELECTRE and Fuzzy TOPSIS methods. All calculation in this thesis were done by using excel programme. The steps of the calculations were explained in the fuzzy methods section. The ranking of the alternatives were calculated different in each method. The main reason behind it was that the calculations were done with different data. Also the steps of each method were different than others. Therefore, finding different solutions were usual.As a result, in order to use in supplier selection problems in construction industry, seven multi-criteria decision making techniques were explained step by step in this thesis. A case study were solved by using these techniques. In further studies, different methods can be used and comparisons between all methods can be done for more accurate results.