A Statistical Synthetic Population Calibration for Activity-Based Model with Incomplete Census Data

摘要: Synthetic population generator is the core component of microsimulation in activity-based travel demand model. Typically, synthetic used way that their decisions on activity-travel pattern are simulated. Traditionally, household sample survey data to synthesize population. The estimated results can be biased due such as low-sampling size and inaccurate data. To deal with this issue, a statistical maximum-likelihood method calibrate using roadside observations (link counts) proposed. Statistical performances proposed evaluated illustrative network real census Multi-day link counts simulated from (true) model parameters Tests carried out assuming different numbers observation variations. illustrate efficiency calibration based its potential for large complex applications.