Mode Choice Estimation for Small Areas

The traditional approach is examined as one of the alternatives and it is concluded that it may be accepted as valid only if a significant change is not expected or desired. The other is the radical approach, which sets a desired level of transit usage. The third alternative is the compromise approach. The objective of this approach is to evaluate the various possibilities about the future transit usage and develop a model that would consider each possibility in proportion to its merit or likelihood, or both. None of the existing techniques used in conjunction with the compromise approach are satisfactory since they lack a rigorous, and yet flexible, analytical basis. The paper presented the use of Bayesian statistics, which explicitly recognizes the uncertainty associated with an estimate, to combine information from various sources, and recommended the technique as it provides the scope of continued surveillance and revision of a model.