Services on Demand
Gestão & Produção
Print version ISSN 0104-530X
On-line version ISSN 1806-9649
SIQUEIRA, Paulo Henrique; CARNIERI, Celso; STEINER, Maria Teresinha Arns and BARBOZA, Ângela Olandoski. A proposed solution for bus driver and fare collector scheduling problems using the maximum weight matching algorithm. Gest. Prod. [online]. 2004, vol.11, n.2, pp.187-196. ISSN 0104-530X. http://dx.doi.org/10.1590/S0104-530X2004000200005.
The purpose of this paper is to discuss how the maximum weight Matching Algorithm can be applied to schedule the workdays of bus drivers and bus fare collectors. This scheduling should be based on the best possible use of timetables in order to minimize the number of employees, overtime and idle hours, thereby minimizing the operational costs of public transportation companies. In the first phase of this study, assuming that the timetables are already divided into long and short duration schedules, the short schedules can be combined to make up an employee's workday. This combination is done by the maximum weight Matching Algorithm, in which the scales are represented by vertices on a graph and the maximum weight is attributed to combinations of scales that do not lead to overtime or idle hours. In the second phase, a weekend schedule is assigned for every weekly work schedule. Based on these two phases, the weekly work schedules of bus drivers and bus fare collectors can be arranged at a minimal cost. The third and final phase of this study consisted of assigning a weekly work schedule to each bus driver and collector, considering his/her preferences. The maximum weight Matching Algorithm was also used in this phase. This method was applied in three public transportation companies in Curitiba, state of Paraná, which had until then used old heuristic algorithms based solely on managerial experience.
Keywords : schedule optimization; maximum weight Matching Algorithm.