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.
