Enhancing Crowdsourcing Through Skill and Willingness-aligned Task Assignment with Workforce Composition Balance

Abstract

Crowdsourcing platforms face critical challenges in task assignment and workforce retention, particularly in aligning complex, skill-intensive tasks with crowd-worker willingness and potential while ensuring workforce diversity and balanced composition. This study introduces the Skill-Aligned Task Assignment and Potential-Aware Workforce Composition (SATA-PAW) framework to address these challenges. The proposed framework formulates the Task Assignment with Workforce Composition Balance (TACOMB) problem as a multi-constraint optimization task, aiming to maximize net utility under task budget constraints while promoting balanced workforce composition. SATA-PAW integrates two novel algorithms, Skill-Aligned Task Assignment (SATA), which optimizes task-worker matching by considering skills, willingness, and budget constraints, and Potential-Aware Workforce Composition (PAW), which leverages satisfaction score and latent potential to retain skilled workers and improve workforce diversity. Experimental evaluations on real-world (UpWork) and synthetic datasets demonstrate SATA-PAW's superiority over five state-of-the-art methods. The results highlight SATA-PAW's ability to integrate human-centric factors with efficient optimization, setting a new benchmark for skill-oriented task assignment and balanced workforce composition in crowdsourcing systems.

Department(s)

Computer Science

Keywords and Phrases

Crowdsourcing; Potential level; Skill-oriented; Task assignment; Willingness; Workforce composition balance

International Standard Serial Number (ISSN)

1574-1192

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2025 Elsevier, All rights reserved.

Publication Date

01 Feb 2025

Link to Full Text

Share

 
COinS