Tvarkaraščio sudarymo bibliotekų tyrimas /

Scheduling problems are common problems that we face when we want to distribute available resources in a limited span of time. These tasks are often solved manually, however complex scheduling tasks are solved only with high computing power. Computers have been used to help solve this problem for more than fifty years. First mention of this can be found in C. Gotlieb's research paper on school scheduling task automation. The aim of this thesis is to analyze the possibilities of one of the selected libraries used for solving planning problems ("JacOp", "Optaplanner", "Choco Solver") based on the restriction groups used in these tasks. The chosen library is tested by solving the university timetabling tasks. Firstly, all commonly used to plan problem restrictions are split into groups. Then all the tasks used in the experiments are designed. Tasks defined for the experiments are each using a different kind of restriction group. Every task is done multiple times using every one of the following algorithms: Tabu search, Simulated Annealing, Late Acceptance Hill Climbing and simple Hill Climbing. The results of this research can be used to help choose the best algorithm for solving a planning problem based on the restriction groups used to define it. Experiments have shown a correlation between different restriction groups and algorithm effectiveness. However the results are not precise enough to state the most effective algorithm for every situation. Therefore more experiments need to be done to get more precise data.

Grantor dc:publisher

Institutional Repository of Vilnius University

Year dc:date

2019

Author dc:creator

• Misiūnas, Simonas,

Contributors dc:contributor

• Birštunas, Adomas

Statement dc:rights

• info:eu-repo/semantics/openAccess

Language dc:language

lit

Repository record dc:identifier

https://repository.vu.lt/VU:ELABAETD81918815&prefLang=en_US

OAI identifier oai:identifier

oai:vu.lt:elaba:81918815