A suggestion for project selection for a small sized contractor through the use of integrated multiple criteria decision making models

Construction sector has become an indispensable sector in Turkey due to its ability to provide great employment and income to people working in this field. The significance of this sector has increased a considerable amount in the recent years following the devastating earthquake in 1999 in Marmara Region, Turkey. Depending on the many studies that were done in order to prevent the harms of potential earthquakes, urban renewal was one of the ideas which was seen as a precaution to this situation. Within the scope of urban renewal, existing buildings with detected damages have been being reconstructed with the aim of building houses that can survive earthquakes with the minimum harm possible. Due to this act, there has been a great increase in the field of construction which directly lead to increase in need of construction firms that will undertake these construction projects. As a consequence of many buildings which are expecting to be reconstructed, construction firms have a lot of different options to choose from. The decision of which project to make an offer is one important step leading to construction firms' success or failure. Moreover, this decision making process is usually complex due to multiple conflicting criteria which have to be considered. Most of the firms make this decision based on their experience and feelings but a decision that is significant like this should also be supported with mathematical methods. In order to help decision makers, many multiple criteria decision making models have been developed in the past. This study is based on a small sized construction firm's decision making process. The purpose of the study is to determine the best project to make an offer among the 12 project alternatives which will be reconstructed in Istanbul through the use of multiple criteria decision making models. Firstly, a literature review was made on the subject of deciding whether to make an offer or not. Depending on the findings, some main criteria that have been found to be effective on this decision were explained and the methods used while making this decision were mentioned in the literature review section of this thesis. Following the literature review section, the multiple criteria decision making methods that are used in this study were explained step by step. The models that are used are Analytical Hierarchy Process (AHP), Preference Ranking Organization Method for Enrichment of Evaluations (PROMETHEE) and Evidential Reasoning. After explaining the methods that are used, the implementation of each of these methods to the decision making process of the construction firm was explained. Before implementing these methods, an interview with the 3 authorized and experienced civil engineers working in the firm were done in order to find out how they make their decisions while making offers. As a result, it was seen that they had never used any mathematical methods but had always made their decisions based on their thoughts and experiences from the past projects. With this interview and also with the help of literature review, the main criteria and subcriteria that are considered by the firm while making this decision were determined. These criteria were determined as company related criteria, project related criteria, cost related criteria, contract related criteria, profit related criteria, management capability related criteria and finance related criteria. First method that was used is AHP. This method is based on the hierarchy that includes the main goal in the first level, criteria in the second, subcriteria in the third and alternatives in the fourth. The pairwise comparisons that AHP requests were made among sequential levels by each of the decision makers. As a result of these comparisons, the importance weights of the criteria, subcriteria and alternatives were calculated. In order to create a common result, geometric mean method was used to combine three evaluations and as a result the ranking of the alternatives was determined. All calculations were done through Microsoft Excel as AHP did not require any additional software. The second method that was used is PROMETHEE. This method required a software called Visual PROMETHEE which can be downloaded for free. PROMETHEE is based on the multiple criteria evaluation tables which are formed separately by three decision makers to evaluate all aternatives depending on each subcriteria. The additional information that PROMETHEE requires are as following: Weights of the subcriteria obtained from AHP, preference functions, indifference and preference values determined by decision makers. Geometric mean method was used in combining the evaluations in evaluation tables like in AHP. As a result of all information that was entered to the software, ranking of the alternatives was obtained. The results were also supported and explained with graphics due to the property of the software. Third method that was used is Evidential Reasoning. This method was applied through a software called Intelligent Decision System that can be downloaded for free. Evidential Reasoning method required only the values in the evaluation tables which were also used for PROMETHEE and the weights of the criteria and sub criteria obtained from AHP. With this method, not just the geometric mean results but each of the values in each of the evaluation tables were included in the software. As a result, the ranking of the alternatives was obtained. Similar to the Visual PROMETHEE software, Intelligent Decision System also provides a graphical demonstration of the results. After the application of all the methods, a summary of the procedure was given in the last section of this thesis. The characteristics of each method was shared and interpreted. Some of the observations are as following: The AHP method was found easy to understand as it did not require any complicated mathematical operations but only simple steps that everyone could easily make. However, the fact that this method requires pairwise comparisons among each sequential level in the hierarchy is considered a disadvantage as it takes a great amount of time which often leads to boredom of decision makers. Another disadvantage is that all pairwise comparisons of the alternatives have to be remade when an additional alternative is to be added or the value of an alternative is to be changed. The PROMETHEE method was found easy to use as the Visual PROMETHEE software did most of the calculations. The indifference and preference values that were included in the solution were easily determined by the decision makers. Moveover, these values made the decision makers feel safer about the decision since they directly could determine their limits through these parameters. Another advantage of this method is that the graphical demonstration the software presents made the results easy to understand and interpret. Evidental Reasoning method was advantageous in the way that it takes each of the three different evaluations into account but not the geometric mean as AHP and PROMETHEE does. It also took less time to apply this method since it does not require any additional information other than the evaluation tables and weights found with AHP. Like in Visual PROMETHEE the graphical demonstration in Intelligent Decision System helped the decision makers to view and interpret the results better. Consequently, it was seen that different results have been obtained from each decision making method as they all required different information and used different mathematical operations to solve the problem. However, it was also seen that PROMETHEE's and Evidential Reasoning methods' solutions were very similar to each other. Similar studies like the one in this thesis should be made in the future in order to make more accurate interpretations among the methods.