The Four Step Model

摘要: The history of demand modeling for person travel has been dominated by the approach that come to be referred as four step model (FSM) (see Chapter 2). Travel, always viewed in theory derived from activity participation, practice modeled with trip-based rather than activity-based methods (as presented 4). Trip origin-destination (O-D) surveys form principle database. influence characteristics decreases, and trip increases, conventional forecasting sequence proceeds. application this is near universal, large measure are its criticisms (these inadequacies well documented, e.g., McNally Recker (1986)). current FSM might best two stages. In first stage, various traveler land use - system (and a varying degree, transportation system) "evaluated, calibrated, validated" produce non-equilibrated (or tables). second loaded onto network process amounts formal equilibration route choice only, not other dimensions such destination, mode, time-of-day, or whether at all (feedback prior stages often introduced, but consistent convergent manner). Although moderately successful aggregate, it failed perform most relevant policy tests, on supply side.This chapter extends material 2 providing concise overview mechanics FSM, illustrated hypothetical case study. discussion chapter, however, will focus U.S. practice. Transportation developed component analysis came established U.S.A. during era post-war development economic growth. Initial analytical began 1950s. landmark study Mitchell Rapkin (1954) only link activities use) called comprehensive framework inquiries into behavior. initial models generation, distribution, diversion early 1950s lead four-step Chicago Area Study Weiner, 1997) sandwiched projection evaluation. was decidedly highway-oriented new facilities being evaluated versus traffic engineering improvements. 1960s brought federal legislation requiring "continuous, comprehensive, cooperative" urban planning, fully institutionalizing FSM. Further 1970s environmental concerns planning modeling, need multimodal planning. It recognized existing may appropriate emerging and, what "first improvement program", call improved led research disaggregate equilibrium assignment integrated have greatly directed approaches last 30 years. late "quick response" (Sosslau et al., 1978; Martin McGuckin, 1998) independently start grown become growing recognition misfit questions 1980s (second, formal) Travel Model Improvement Program 1991; much subsequent period improving state-of-thepractice relative while fostering methodologies further state-of-the-art 4).The seen particular systems (TSA), due Manheim (1979) Florian al. (1988), which positions view strengths weaknesses. A brief presentation TSA introduces context leads problem area definition, application, data requirements. perhaps commonly utilized then sample application.