Monte Carlo simulation of household travel survey data with Bayesian updating

Peter R. Stopher, Graham Pointer

Research output: Contribution to journalReview articlepeer-review

3 Scopus citations


Several recent papers have demonstrated the feasibility of using Monte Carlo simulation to simulate characteristics that would be collected in a household travel survey - number of trips by purpose, mode of travel, trip length, and time of departure - using sociodemographic data on the household from a census, and distributions of travel characteristics from a national travel survey. Although shown to be feasible as a method, the results tended to reflect still more of the nature of the source national travel data, and not as much of the local characteristics as would be desired. This paper describes research in which the distributions of travel characteristics were updated from a small local sample, using the method of Bayesian updating with subjective priors. The resulting distributions were found to resemble more closely those of the local area for which the simulation was being undertaken. The paper describes the procedure of updating. It then summarises the results obtained in updating the distributions for Adelaide in South Australia, using original distributions from the US and local census data, together with a local household travel survey. The paper also shows the results of expanding the simulated data to the region. These results are compared with the expansion of actual survey data for the region. It is found that the updated simulation gives a more accurate simulation result for the region than simulation without updating. This procedure offers a low-cost method for regions to develop simulated household travel survey data that can be used for description of the region's travel and the development of travel forecasting models, but at costs that are a fraction of those associated with normal household travel surveys. The updating procedure requires a local sample of about 500 households, and the simulation can produce as large a sample as may be desired for the region - up to the entire population - but at a cost of a few person hours of work.

Original languageEnglish (US)
Pages (from-to)22-32
Number of pages11
JournalRoad and Transport Research
Issue number4
StatePublished - Dec 2004
Externally publishedYes

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Mechanical Engineering


Dive into the research topics of 'Monte Carlo simulation of household travel survey data with Bayesian updating'. Together they form a unique fingerprint.

Cite this