چكيده به لاتين
The importance of maintenance of any system, which its components are worn during
operation, is clear. Railway transportation system as a global and public system, requires
permanent monitoring, maintenance. Because its reliability, safety and efficiency are
highly dependent on the quality of the railway tracks. Due to the fact that the length of
the railway tracks and the required maintenance volume are very high, the use of human
resources to implement projects is uneconomic and virtually impossible. This is done by
mechanized machinery used on railways. Because of the limited number of machines and
the need to implement projects as many as possible, it is necessary to develop a heuristic
method capable to take both of the objectives and the priorities and requirements of the
railway into account, at the same time.
In this thesis, the scheduling process of Iran railway tracks preventive maintenance
operation has been studied. Since this is a NP-hard problem, the use of heuristics and
meta-heuristics can be very useful. A heuristic algorithm is designed to apply all of the
railway priorities and requirements. Due to the local consideration applied in this solution
method, a hybrid algorithm is also proposed. This algorithm involves two levels of
making decision in the order of input projects and the choice of the best machine for
scheduling. At the first level, the Tabu Search algorithm and at the second level, the Ant
Colony System (ACS) have been used. In order to have a more efficient implementation
of the algorithm and to consider the most important indicators as the objective function,
the algorithm is designed in two phases. In the first phase, the total scheduled length (km)
is to be maximized. In the second phase, the objective is to prevent the deterioration of
the first phase objective and to minimize the machines travel. This algorithm could
generate better results compared with the manual plan prepared by the Iran railways. The
simulation of the real world conditions of Iran railways has been used for better evaluation
of the prepared timetable. Furthermore, the use of simulation for the runtime of projects
rather than using normal calculations has been examined. The results show that
uncertainty has an important impact on the implementation and the time of the activities,
which is seen in simulation.
