An adaptive approach for solving dynamic scheduling with time-varying number of tasks — Part I

Abstract

Changes in environment is common in daily activities and usually introduce new problems. To be adaptive to these changes, new solutions to the problems are to be found every time change occur.Our previous publication showed that centroid of nondominated solutions associated with Multi-Objective Evolutionary Algorithm (MOEA) from previous changes enhances the search quality of solutions for the current change. However, the number of tasks in the test environment employed was fixed. In this two-part paper, we address the dynamic adaptation with time-varying task number.To cope with this variability, new components of the solution, corresponding to the new tasks, are inserted appropriately to all solutions of the previous changes. Then centroid of these modified solutions is recomputed. Further, to avoid confusion in solution presentation, the insertion of new tasks obliged the use of task ID number greater than the largest of the previous IDs. The first part of this paper will show that the resulting task numbering system will alter the centroid significantly which will degrade MOEA's search quality. To circumvent, task IDs are mapped to new values in order to minimize difference in IDs between adjacent solution components; an approach which significantly upgraded the search performance despite changes in task number as supported by the obtained results.

Multiobjective Resource-Constrained Project Scheduling with a Time-Varying Number of Tasks

“…The insertion side effect was resolved in [14] by mapping IDs of tasks, that belong to a similar precedence order, to unique values that are as close to each other as possible. Let the function F ( I d ) represent the ID-mapping operation, where I d is the ID to be mapped.…”

“…Legitimization of a subalgorithm of a technique is to manifest the decline in the effectiveness of this technique in solving a problem when the subalgorithm is replaced;extend the investigations in our previous work [14]. In particular, add the techniques being compared to McBAR with the technique that utilizes Estimation Distribution Algorithm (EDA) [15].…”

“…Our previous work [14] investigated the performance (solution-searching ability) of the memory and EA-based technique called mapping of task IDs for centroid-based adaptation with random immigrants (McBAR) in solving the Φ 2 problem. However, this investigation was performed on few instances of Φ 2 .…”

“…Let Φ 2 denote the biobjective dynamic RCPS problem where minimization of the schedule cost and duration are the conflicting objectives and where the number of tasks varies in time, a variation that brings about a change in the dimension of this problem. Our previous work [ 14 ] investigated the performance (solution-searching ability) of the memory and EA-based technique called mapping of task IDs for centroid-based adaptation with random immigrants (McBAR) in solving the Φ 2 problem. However, this investigation was performed on few instances of Φ 2 .…”

A Hybrid Predictive Strategy Carried through Simultaneously from Decision Space and Objective Space for Evolutionary Dynamic Multiobjective Optimization

“…In the recent literature, whether in industrial applications or scientific research, there is a lot of contents related to DMOPs [1][2][3]. Industrial applications involve design [4,5], management [6,7], scheduling [8][9][10], planning [11][12][13][14], and control [15,16]. Scientific research includes constrained optimization [17,18], machine learning [19,20], and bilevel optimization [21].…”