Task Decomposition Model for Dispatchers in Dynanic Scheduling of Demand Responsive Transit Systems

Rahimi, Mansour; Dessouky, Maged; Gounaris, Ioannis; Placencia, Greg; Weidner, Merrill · 2000 · ROSA P / University of Southern California. Department of Industrial and Systems Engineering

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study addresses the lack of research regarding the cognitive complexity and task requirements of dispatchers in demand-responsive paratransit systems, particularly following the increased demand driven by the Americans with Disabilities Act (ADA). While paratransit providers increasingly rely on computer automation to streamline operations, little is known about how dispatchers interact with these systems or the specific challenges they face. The research aims to fill this gap by developing a task analytic model of dispatcher activities and evaluating the human-computer interaction (HCI) aspects of current dispatching software. The goal is to provide a formalized understanding of dispatcher tasks to improve training methodologies and guide future software interface design. The researchers conducted an in-depth analysis of a Los Angeles-area paratransit service provider, referred to as "ABC." The methodology involved extensive field observations, video recordings, and interviews with expert dispatchers to capture task sequences during high-workload periods. Using Hierarchical Task Analysis (HTA), the team refined these observations into a comprehensive decision hierarchy, or decision tree, that maps the sequence of activities and cognitive decisions made by both van and lead dispatchers. Additionally, the study evaluated two software interfaces used in the operation: a DOS-based system and a Windows-based graphical user interface (GUI) with GPS capabilities. The analysis focused on identifying bottlenecks, cognitive loads, and usability issues within these systems. The findings reveal that dispatcher tasks involve intensive and complex cognitive processes heavily influenced by software interface design. The HTA model successfully captured these interactions, demonstrating its utility as a tool for part-task training for entry-level dispatchers. In the HCI evaluation, the DOS-based system was found to offer information simplicity but suffered from long scanning and navigation times, inconsistent screen designs, and a steep learning curve. Conversely, the Windows-based system provided more natural spatial representations and efficient popup menus but was critically flawed by significant system lag times, sometimes reaching 30 seconds. This latency prevented dispatchers from using the Windows system during high-demand periods. Other disadvantages of the Windows interface included high-density visual clutter, layout inconsistencies, inappropriate color-coding, and a lack of direct driver interaction during GPS failures. The significance of this research lies in its recommendation for a redesign of dispatch software based on user-centered interface design principles. The authors argue that current systems fail to support the rapid decision-making required in dynamic scheduling, leading to errors and reduced productivity. By utilizing the developed HTA model, agencies can better train dispatchers and identify specific interface weaknesses. The study concludes that software developers must prioritize natural, quick-response interfaces to effectively support the complex cognitive demands of paratransit dispatching, ultimately improving service efficiency and reducing operator stress.

Key finding

Dispatchers abandoned the Windows-based system during high-demand periods due to a system lag time of up to 30 seconds, despite its superior spatial representation compared to the DOS-based system.

Methodology

field_study

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• mental demand