Multi-Objective Task Scheduling Optimization in Spatial Crowdsourcing

Abstract

Recently, with the development of mobile devices and the crowdsourcing platform, spatial crowdsourcing (SC) has become more widespread. In SC, workers need to physically travel to complete spatial–temporal tasks during a certain period of time. The main problem in SC platforms is scheduling a set of proper workers to achieve a set of spatial tasks based on different objectives. In actuality, real-world applications of SC need to optimize multiple objectives together, and these objectives may sometimes conflict with one another. Furthermore, there is a lack of research dealing with the multi-objective optimization (MOO) problem within an SC environment. Thus, in this work we focused on task scheduling based on multi-objective optimization (TS-MOO) in SC, which is based on maximizing the number of completed tasks, minimizing the total travel costs, and ensuring the balance of the workload between workers. To solve the previous problem, we developed a new method, i.e., the multi-objective task scheduling optimization (MOTSO) model that consists of two algorithms, namely, the multi-objective particle swarm optimization (MOPSO) algorithm with our fitness function Alabbadi, et al. and the ranking strategy algorithm based on the task entropy concept and task execution duration. The main purpose of our ranking strategy is to improve and enhance the performance of our MOPSO. The primary goal of the proposed MOTSO model is to find an optimal solution based on the multiple objectives that conflict with one another. We conducted our experiment with both synthetic and real datasets; the experimental results and statistical analysis showed that our proposed model is effective in terms of maximizing the number of completed tasks, minimizing the total travel costs, and balancing the workload between workers.

MOTS‐ACO: An improved ant colony optimiser for multi‐objective task scheduling optimisation problem in cloud data centres

“…The contribution of the study shows that the proposed scheduling approach makes the task mapping perform in a better manner. Alabbadi and Abulkhair [34] proposed a novel Multi‐Objective Particle Swarm Optimisation (MOPSO) algorithm for multi‐objective task scheduling on cloud data centres, and the results showed that MOPSO can find the optimal solution and outperform other existing single and multi‐objective scheduling algorithms. Generally, previous studies only focussed on makespan [35, 36] and power usage [37, 38] in the scheduling process, or only considered the load balancing [39, 40], and employed certain algorithms to achieve multi‐objective [41, 42] or single‐objective optimisation.…”

Research on Agricultural Machinery Rental Optimization Based on the Dynamic Artificial Bee-Ant Colony Algorithm

“…Kumar et al proposed a generalized ant colony optimizer algorithm (GACO) to solve the cloud resource allocation problem [14]. Alabbadi et al proposed multi-objective task scheduling optimization (MOTSO) to find the optimal solution [15], which solved the optimization problem well. Cao et al proposed a depth optimization analysis method based on the GA algorithm [16].…”