University Links: Home Page | Site Map
Covenant University Repository

Machine Learning Priority Rule (MLPR) For Solving Resource-Constrained Project Scheduling Problems

Adamu, Patience I. and Aromolaran, Olufemi (2018) Machine Learning Priority Rule (MLPR) For Solving Resource-Constrained Project Scheduling Problems. In: International MultiConference of Engineers and Computer Scientists, March 14-16, 2018, Hong Kong.

[img] PDF
Download (1MB)


This paper introduces a machine learning priority rule for solving non-preemptive resource-constrained project scheduling problems (RCPSP). The objective is to find a schedule of the project’s tasks that minimizes the total completion time of the project satisfying the precedence and resource constraints. Priority rule based scheduling technique is a scheduling method for constructing feasible schedules of the jobs of projects. This approach is made up of two parts: a priority rule to determine the activity list and a schedule generation scheme which constructs the feasible schedule of the constructed activity list. Different scheduling methods use one of these schemes to construct schedules to obtain the overall project completion time. Quite a number of priority rules are available; selecting the best one for a particular input problem is extremely difficult. We present a machine learning priority rule which assembles a set of priority rules, and uses machine learning strategies to choose the one with the best performance at every point in time to construct an activity list of a project. The one with better performance is used most frequently. This removes the problem of manually searching for the best priority rule amongst the dozens of rules that are available. We used our approach to solve a fictitious project with 11 activities from Pm Knowledge Center. Four priority rules were combined. We used serial schedule generation scheme to generate our schedules. Our result showed that the total completion time of the project obtained with our approach competes favorably with the completion times gotten with the component priority rules. We then went further and compared our algorithm with 9 other available priority rules. Our results showed that the completion time got using our algorithm compete favorably with the total 13 priority rules available in the literature.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: machine learning, motion planning, network analysis, resource constraints, probabilistic roadmap planners
Subjects: Q Science > QA Mathematics
Divisions: Faculty of Engineering, Science and Mathematics > School of Mathematics
Depositing User: Mrs Patricia Nwokealisi
Date Deposited: 26 Jun 2018 09:56
Last Modified: 26 Jun 2018 09:56

Actions (login required)

View Item View Item