The Interleaved Constructive Memetic Algorithm and its application to timetabling

Abstract

Timetabling problems are well known NP-hard constraint satisfaction, and real-world cases often have complicated and challenging structures. For such problems, we present a new hybrid method, "Interleaved Constructive Memetic Algorithm" (ICMA) that interleaves memetic algorithms with constructive methods. ICMA works using an active subset of all the events. Starting with a few events, in multiple construction stages ICMA increases the active set to eventually include all of them. At each stage, a memetic algorithm (MA) is applied to improve the current partial solution before the next construction step. We also describe a real-world course timetabling problem, the "Preparation School Timetabling Problem" (PSTP), which is of particular interest because it has a highly hierarchical structure arising from various organisational requirements. An important advantage of ICMA is that both the constructive heuristics and MA can be tailored to exploit such hierarchical structures. We give empirical results showing that ICMA performs better than the corresponding conventional MA on the PSTP.

Modifying Regeneration Mutation and Hybridising Clonal Selection for Evolutionary Algorithms Based Timetabling Tool

“…MAs have received considerable attention from researchers in many fields [5] including job shop scheduling [16], vehicle routing [17], exam timetabling [18], and nurse scheduling [19]. The MA has also been applied to solve course timetabling problems [20][21][22][23][24][25].…”

A self-adaptive Multimeme Memetic Algorithm co-evolving utility scores to control genetic operators and their parameter settings

“…A simple MA introduces a local search phase into a GA after crossover and mutation have been performed during the evolutionary process. Since their emergence, MAs and subsequent variants of MAs have been applied to a wide variety of problems, including educational timetabling [2,10,56,60,59], multi-objective optimisation problems [31,24], permanent-magnet synchronous motor (PMSM) design [11], Optimal Placement of PMUs [37], permutation flow shop scheduling [28], the economic load dispatch problem [51,52], protein folding [33], HIV therapy design [44], the quadratic assignment problem [39] and the travelling salesman problem [42,5,26]. As well as the application of MAs to practical optimisation problems, a number of studies have sought to understand the concepts underpinning MAs and the behaviour of memes [34].…”

Memetic algorithms for Cross-domain Heuristic Search

“…Memetic algorithms (MAs), also referred to as hybrid genetic algorithms, represent a set of genetic algorithms that make heavy use of hill climbing. Ever since its emergence, memetic algorithms and subsequent variants of MAs have been applied to various problems such as the timetabling problems [7], [1], [42], [37], [9], [43], permutation flow shop scheduling [21], travelling salesman problem [31], [2], [16], quadratic assignment problem [28], multi-objective optimization problems [23], [11] and protein folding problem [25]. Apart from application of memetic algorithms on various problem domains, different meme considerations has also been well studied.…”