Search with evolutionary ruin and stochastic rebuild: A theoretic framework and a case study on exam timetabling

Abstract

-This paper presents a new search method, called Evolutionary Ruin and Stochastic Recreate, which tries to learn and adapt to the changing environments during the search process. It improves the performance of the original Ruin and Recreate principle by embedding an additional phase of Evolutionary Ruin to mimic the evolution within single solutions. This method executes a cycle of Solution Decomposition, Evolutionary Ruin, Stochastic Recreate and Solution Acceptance until a certain stopping condition is met. The Solution Decomposition phase first uses some problem-specific knowledge to decompose a solution into its components and assign a score to each. The Evolutionary Ruin phase then employs two evolutionary operators (namely Selection and Mutation) to destroy a certain fraction of the entire solution. After the above phases, an input complete solution becomes partial and thus the resulting partial solution needs to be repaired. The repair is done by the Stochastic Recreate phase which reintroduces the removed items in a specific way and asks the underlying improvement heuristic whether this move should be accepted or not. Last, the Solution Acceptance phase selects a specific strategy to determine the probability of accepting the newly generated solution. Hence, optimisation is achieved by an iterative process of component evaluation, solution disruption and stochastic constructive repair. The well-known exam timetabling problems are used to test the availability of the approach in solving real-world hard problems. Furthermore, this paper presents a formal framework and implements a Markov chain analysis on some theoretical properties of the approach.

Shift Scheduling and Employee Rostering: An Evolutionary Ruin & Stochastic Recreate Solution

“…To solve this problem, we apply ER&SR first defined in [11] then explored as a theoretical framework in [10]. ER&SR is an algorithm derived from the traditional Ruin and Recreate algorithm defined in [16].…”

“…ER&SR is an algorithm derived from the traditional Ruin and Recreate algorithm defined in [16]. This methodology was chosen as it has thus far been used to solve a complex exam timetabling problem in [10]. More research is necessary in its application in similar problem areas, employee scheduling & rostering being one such area.…”

“…The rest of this paper is structured as follows: in Section 2 the problem description is defined as are the prospective benefits of this approach. Section 3 details the solution implementation in Java, using the theoretical framework provided by [10] as a basis, and the unique elements of this implementation are highlighted. The results of using this approach are presented and discussed in Section 4.…”

A hybrid Integer Programming and Variable Neighbourhood Search algorithm to solve Nurse Rostering Problems

“…[27] have also used this strategy by the evolutionary elimination of parts of the solution and subsequently repairing it by using a greedy heuristic. A more advanced ruin-and-recreate based algorithm is also reported in [28], where the authors applied a stochastic modelling and Markov chain analysis. Nonetheless, in the proposed hybrid algorithm, we apply this strategy not only for the diversification purpose but also for improving the quality of the obtained solution, i.e.…”

Variable Neighbourhood Search: A case study for a highly-constrained workforce scheduling problem

“…Demand forecasting being one such suggested topic for further study, which we have implemented in this case study. Multiple fields have benefited from the use of metaheuristics to solve large complex rostering scheduling problems, including nurse rostering [3,4,5], truck scheduling [6] and exam timetabling [7,8]. It goes to follow that a similarly highly constrained search optimization problem in engineer rostering can benefit from similar techniques.…”