The recommendation of a decision support system for maintenance management: An application for the public transportation process

All equipment in a production or service enterprise has various failures due to aging and wear over time. Maintenance activities can be detemined as checking equipment before failure or after repair. Maintenance activities are to check the equipment before failure or to perform repair after failure. It is important to determine an effective policy by addressing the stochastic uncertainty of failures in maintenance activities and the fuzzy uncertainty associated with human intervention. In this context, the behavior of previously experienced failures can be analyzed and a rule based maintenance decision-making mechanism can be developed. With this mechanism, the appropriate instant maintenance decision on the vehicles can be given easily and effectively. The main purpose of maintenance activities is to ensure the efficient use of all equipment and to prevent unexpected failures. Failures are unpredictable and show stochastic uncertainty. Moreover, due to the greater inclusion of human factor in service businesses and the unpredictable nature of this factor, fuzzy uncertainty as well as stochastic uncertainty in maintenance activities should be considered. In recent years, increasing urbanization and transportation problems in parallel with the rapidly rising population in cities are emerging. The steps to be taken to solve these problems are not only new vehicles, alternatives or paths to be added to public transport systems; but also to carry out a more planned operation to improve the service quality of existing vehicles, and to perform maintenance and control. Public transportation vehicles are exposed to wear due to age and use. This leads to a decline in service quality, increased passenger dissatisfaction and costs, and even the emergence of situations that endanger the lives of passengers. This has led to the shift of interest to the determination of the maintenance policy, which prevents the failure of these systems, reduces maintenance costs and increases system reliability. Besides, it is seen in the literature that the most important problem affecting the cost of enterprises is the creation of an optimum maintenance model. In this sense, pre-determination of failures, reliability calculations and maintenance policy decision are very important for firms operating in the production sector. However, these issues are of great importance for the companies that produce services such as transportation sector for the reasons mentioned above. The Metrobus system, introduced in 2007, is an important transportation line which increases the traffic density in Istanbul and increases the interest in public transportation. The metrobus system offers daily transportation to approximately 715.000 passengers between Beylikdüzü-Söğütlüçeşme. Therefore, it is clear that a large number of people will be affected by any failure & stop event occurring in this system. In this context, it is very important to create an effective maintenance policy and to develop an inference decision mechanism for randomly generated failures. In this thesis, a rule-based inference system has been developed for failures in the BRT system. Fuzzy logic is used to analyze uncertainties in the process. Factors expressed as input variables with different sizes such as vehicle age, effect of failures, cost, reliability, driver ability have been determined and a decision support system has been established for actions such as renewal, maintenance or continued use of metrobus vehicles. In the prioritization study to determine the effects of failures, a significant contribution is made by using the stochastic MCDM, because it is the first study on the literature both in terms of method and in the relevant field. As a result, a system has been developed in which tools such as resumption of instant use or maintenance can be easily given. A system has been developed in which the decision to use or maintenance can be easily given instantly on the basis of every vehicles.