Over the last few decades, competition among maritime container terminals has increased due to large growth rates on major seaborne container routes. Thus terminals have to carefully plan the management of their facilities and services in order to satisfy shipping companies’ demands. Failure to do so will lead customers to re-schedule their routes calling at new promising terminals.As a result, and so as not to lose competitiveness, container terminals are showing a growing interest in advanced decision support systems for their management, in which optimization methods play a crucial role.In this context, the optimal management of human resources is a major issue for terminals, particularly in case of high labor costs. Workload is organized in shifts covering 24 hours a day. Due to union and work rules, shifts are typically required to be planned a number of months in advance. However, a significant feature of terminal activities is the lack of precise information about the time each activity will be performed. This problem should be addressed by greater flexibility, which can be achieved by defining two planning levels: long term, which uses a monthly horizon and short term (operational), which encompasses one or more days [1, 2].In this work we address the operational planning level, because it inherits some unchangeable decisions from the long term plan, involves the use of terminal resources considering the real work demand and leads to final estimations of benefits deriving from optimization processes. Although there is plenty of literature on workforce management in several areas, little attention has been devoted to the specific context of maritime container terminals . Moreover, none of these formulations have highlighted personnel shortfall, which is the most tricky situation for terminals, because it may result in vessel delays. In these cases, high penalties are charged by shipping companies to terminals.In this work we present an integer linear programming model to address this gap. The objective of the model is to determine the optimal allocation of workers to tasks, shifts and terminal activities, as well as to minimize shortfalls. Different requirements are taken into account for permanent staff and external workers. Moreover, since a longer-than-a-day planning horizon can exploit the visibility of more data, we aim to investigate the effect of a longer planning horizon on this problem. The proposed model is exactly solved by a state-of-art solver within reasonable times for the needs of terminal containers. Preliminary tests show that two-day planning horizons lead to lower shortages than standard daily ones. Moreover, whenever shortfalls are observed, they seem to be clustered in low level tasks, which are easier to tackle by overtime.
|Numero di pagine||2|
|Stato di pubblicazione||Published - 2012|