This paper describes an improved method of estimating the fraction of vehicle-miles of travel (VMT) operating in cold start mode, that uses an enhanced traffic assignment technique. It begins with a literature review; most of the previous work on the analysis of cold starts was done in the 1970s. As there was not yet any standard procedure for estimating the proportion of VMT, operating in cold start mode, a new procedure was developed for air quality conformity analyses, using regional traffic models for Northern and Southern New Jersey, USA. A major difficulty in analysing cold start travel is identifying where a cold start trip changes to stabilised mode. Versions of the MINUTP travel forecasting software package, from April 1992 on, have this capability, which enables the detailed analysis of cold start traffic at the link level. An illustrative example is given of the use of MINUTP; users must provide estimates of the cold start trip proportion and interval duration. Some results of applying the new procedure to South Jersey are tabulated; the total cold start VMT proportion was 18%. The effects of cold start estimation procedures on forecast changes are also tabulated.
Abstract