Modeling Mode and Time-of-Day Choice with Joint RP and SC Data

Trip departure time has become a more important theme in practice as urban congestion problems are increasingly addressed by travel demand management (TDM) strategies. In this paper, the authors formulate and estimate a joint mode-departure time choice model using combining revealed preference (RP) and stated choice (SC) data about commuting trips in Santiago. The information was gathered through a series of surveys (RP, SC and attitudinal survey) applied to some 500 commuters in the Santiago Metropolitan Area. The travel time, cost and cost divided by wage rate coefficients were fairly similar in both environments (RP and SC), while schedule delay (SD) penalties associated with early or late arrival to work differed between each data. The degree of flexibility that workers have to adjust their arrival time to work resulted to be statistically significant when interacted with SD terms, suggesting that the level of work flexibility indeed influences temporal choices. The use of different time-resolution intervals showed that goodness of fit of the estimated models increased when higher time resolutions (i.e. length of departure time intervals) were considered, but values of time could differ when using distinct aggregation of trip departure time alternatives.