The rate of absenteeism averages about 3.5% across all occupations in the US labor force, while in the transit industry absenteeism among operators exceeds 10%. Prior research attributes the transit industry’s higher operator absenteeism to factors falling in three general categories: 1) higher levels of stress; 2) economic; and 3) work rules. Vacancies created by absences are referred to as open work. Open work can be filled from a pool of operators held in reserve, by calling in off-duty operators, and by re-assigning the work as overtime among in-service operators. Research indicates a concern that the latter two options may themselves contribute to subsequent absenteeism. Also, there has been a trend toward adding scheduled overtime in regular work assignments in order to contain total labor costs. Operators with scheduled overtime assignments are less likely to fill open work, occurring while they are in service or during their days off, leading to greater reliance on... View Full Summary

Hide Full Summary
The rate of absenteeism averages about 3.5% across all occupations in the US labor force, while in the transit industry absenteeism among operators exceeds 10%. Prior research attributes the transit industry’s higher operator absenteeism to factors falling in three general categories: 1) higher levels of stress; 2) economic; and 3) work rules. Vacancies created by absences are referred to as open work. Open work can be filled from a pool of operators held in reserve, by calling in off-duty operators, and by re-assigning the work as overtime among in-service operators. Research indicates a concern that the latter two options may themselves contribute to subsequent absenteeism. Also, there has been a trend toward adding scheduled overtime in regular work assignments in order to contain total labor costs. Operators with scheduled overtime assignments are less likely to fill open work, occurring while they are in service or during their days off, leading to greater reliance on the reserve operator pool. If open work cannot be reassigned, missed pull-outs result. Missed pull-outs often occur during peak periods and affect greater numbers of passengers. Thus transit providers seek to minimize missed pull-outs. Work force planning for filling open work is easier when operator absence rates are fairly consistent from day to day. This appears not to be the case in the transit industry. Daily variations in absences means that on some days the amount of open work exceeds operator availability, resulting in missed pull-outs, and on other days operator supply exceeds open work, resulting in payment to unassigned operators (i.e., report time). Both outcomes involve costs, with the former borne by passengers (through longer waits or foregone trips) and the latter borne by the transit provider. Prior research on absenteeism among transit operators has involved empirical analysis of contributing factors within time units that range from week-long to quarter-long periods, which fail to address daily variations. Prior research has also tended to focus on absence history, scheduled hours of work and earnings, and operator demographics, while paying less attention to variables that might proxy for stress associated with the assigned work. The proposed research seeks to analyze daily bus operator absence patterns at TriMet. One advantage of the focus on this transit provider is its detailed archive of operating data from automatic passenger counter (APC) and automatic vehicle location (AVL) technologies, providing a rich source of information related to factor influencing on-the-job stress. Operations data can also be directly linked to operator-specific characteristics included in the agency’s human resources database. The proposed project will analyze daily absenteeism patterns in two ways. First, it will identify the primary dimensions of the variation in daily rates of absenteeism with respect to temporal factors (i.e., weekday v. weekend, day of the week, month/season), service typology (peak/off-peak periods, regular/limited/express service), route typology (trunk/radial, cross-town, feeder), and operator typology (full/part time, experience, demographics). Second, an empirical model of an operator’s daily decision to report for work will be estimated. Determinants of the decision to report for work will include variables related to each operator’s employment and assigned work status, recent operations performance characteristics of their assigned work, and their prior record of attendance. Findings from the analysis may prove useful in identifying changes in policies or practices that would reduce the rate or variance of operator absences. Alternatively, workforce planning practices may benefit from improved understanding of absence patterns among operators.

TriMet, PSU Toulan School of Urban Studies and Planning

Project Type: Research
Start Date: October 1, 2007
End Date: September 30, 2008
Related Projects: None
Research Area: Healthy Communities
RiP Number: 14663