Agent-Based Microsimulation Approach for Design and Evaluation of Flexible Work Schedules

Abstract

The policy of a flexible work schedule has been practiced for years in order to stimulate the redistribution of departure time of commuters. However, its potential influence on travelers’ departure time shifts is infrequently seen in existing studies. This study extended an agent-based positive departure time choice model to gain perspective on travelers’ dynamic reaction toward the flexible work schedule policy. Unlike most rational behavior models, the positive model emphasizes the bounded rationality in people's actual behavior and allows for heterogeneity among travelers. Dynamic traffic assignment is integrated with this proposed model to build up a feedback loop between individual choice (demand side) and network performance (supply side). Scenarios of different percentages of the population with a flexible work schedule are analyzed. It is found that travelers with flexibility in their work schedules tend to depart for work later to avoid peak periods in the morning. The average travel time in the network will decrease by at most 22% when 60% of the travelers have flexible work schedules.