KUniKIP is an optimization tool for the construction of optimized event schedules for the yearly event KinderUni
of the Freie Universität Berlin. For a few days the FU offers a selection of courses with differing dates school classes with children aged between 8 to 12 years
can apply for. These classes are then to be assigned to single course dates according to their preferences. However, the courses usually come with capacity and other
restrictions and can thus not cover the huge number of applications. We therefore desire an "optimal" solution under the given circumstances and custom
parameters to give as many children as possible the chance to participate. KUniKIP allows us to do just that by modeling the multiple knapsack problem.
Herein the courses are interpreted as knapsacks and the classes as the items we want to distribute over the knapsacks. The classes' weight is given by the amount
of pupils in them. The objective function is formulated in respect to what property we seek to maximize, be it the number of pupils, classes, (first-time-)schools or
the number of dates with an assignment. KUniKIP will also, to some extend, analyze the given problem to give more insight in the quality of the solution.
KUniKIP is written in Perl and utilizes the ZIB Optimization Suite. It is available for Windows and Linux OS and should be adaptable
to similar problems. The program and the math behind it are the result of a Diploma thesis
that emerged from a joint project of the Zuse Institute Berlin (ZIB) and the Freie Universität Berlin (FU)
and was improved and refined by several developers after. It is still being maintained at the ZIB and in use by the FU's KinderUni.
