Daily Stop-Making Model for Workers

Abstract

A mathematical model was used to predict workers' stop-making propensity as well as to discern the potential interactions in stop making over different time periods in a worker's day. Five time periods were examined, including before the morning commute, morning commute, midday, evening commute, and after the evening commute. With the New York Metropolitan Transportation Council's household survey data, a multivariate probit model was developed to allow the joint determination of workers' stop-making choices (across different times of day). To establish the stop-making relationship, the model took into account factors relating to sociodemographics, land use density, transportation service, and work schedule. Extensive statistical tests were given to the model specifications and the assumptions underlying the model structure. The possible interdependence of stop-making decisions was analyzed by using the modeling results, including the sign and magnitude of the correlation coefficients. The empirical results provide useful insights into the substitution and complementarity effects in stop making among the five time periods. They also provide exploratory methodological evidence that could lead to an approach to determine whether to assume an independent or a simultaneous choice structure for characterizing the true choice process of workers' stop-making decisions.