Enhanced timetable scheduling: a high-performance computational approach

Abstract

Timetable scheduling is a complicated, expensive and resource-intensive Optimization Problem. This project aims to suggest a solution to this problem using multiple strategies. The core strategy is to use Artificial Intelligence and Machine Learning to optimize a timetable. The result is optimized further by reapplying this optimization mechanism iteratively without aiming to build a perfect result in a single iteration. The project uses the concepts of High-Performance Computing and Cluster Computing to provide flexibility and efficiency on a hardware level. These form the basis of Project Almanac: a robust and flexible timetable optimization architecture. Project Almanac aims to generate a ‘good enough’ timetable by adjusting the expenses according to the end-user requirements. Alternatively, the solution also intends to offer a faster, cheaper and more flexible hardware-software architecture to generate optimized timetables for diverse applications.