Passeport Vacances is an offer for school-aged children to discover a set of activities during holidays. For more than 30 years, it has been an established social function in several countries, including Germany and Switzerland. Proposed activities might occur several times during the Passeport Vacances. The assignment of activities to children is computed in order to maximize the children's preferences, as well as to balance each child's incurred cost, toward an equity goal. There are several sets of constraints associated with the assignment problem: no overlapping activities assigned to the same child, minimal and maximal ages per activity, minimum number of children for opening an activity, maximal size of a group for each activity, no similar activities assigned to the same child, no already assigned 'lifetime'-activity per child, and at most one activity per period and per child. We propose a binary linear programming model that describes the assignment problem, report CPU computation issues regarding the model implementation, and report numerical results based on a state-of-the-art MIP solver. Tests where conducted with real data from the 2016 edition of Passeport Vacances in Morges.

“Passeport Vacances”, abbreviated PV, is a set of leisure activities proposed to children to discover and enjoy during school holidays. During PV, activities are proposed several times, each one being an occurrence. This data set contains real data, collected by online registration during the summer of 2017. Children express their preferences for each available time slot. Organizers should assign activities to children by maximizing their expressed preferences, subject to several types of constraints: age limit, group size limit for each occurrence of an activity, diversification of the type of activities for each child, restrictions on costly activities, restrictions on the number of activities per period, and cost balancing. The CSV files in this data set represent the preferences of 634 children for 1121 activities over a two-week period. These data were used to develop the Morges 2017 Vacation Passport model, which is associated with the research article entitled ““Passeport Vacances”: an assignment problem with cost balancing” Beffa and Varone, 2018.

ÐWe study a production scheduling problem, which adresses on the one hand the usual operational constraints such as the precedence of operations, time windows, delays, uniqueness of treatment, availability of resources, and waiting times. On the other hand, the problem takes into account possible restricted movements according to production orders. This problem is a variant of a flexible job shop scheduling problem with several types of sequence-dependent constraints. We consider additional sequence-dependent setup times, as well as sequence-dependent transportation and assignment restrictions. We propose a mixed integer programming model (MIP). It is based on the MIP model of a flexible job shop scheduling problem, in which we add those sequence-dependent constraints. We solve it with a general purpose MIP solver.