However, the present study infers that the policy of increasing comfort with increase in travel cost of buses will
not help in improving the patronage of buses.
7. CONCLUSION
The main objective of the paper was to test the applicability of fuzzy logic for modeling mode choice, since
fuzzy logic is capable of capturing human reasoning and logic in a much better way than the probabilistic models
that are usually employed for the purpose. For this purpose, Port Blair city was selected and appropriate data
though household survey has been collected. From the analysis of this study, it can be satisfactorily said that,
fuzzy logic technique provides a better performance in analyzing mode choice compared to the traditional
multinomial logit model. It was observed that the multinomial logit model was able to predict only two modes
i.e. bus and two wheeler with an overall accuracy of 40% where as, the fuzzy logic model was able to predict all
modes with a much better overall accuracy of 69%. Amongst the influencing variables, it was noted that invehicle
travel time and travel cost are the biggest determinants of modal choice.
Following this certain policy measures, aimed at increasing the modal share of public transport in the total
trips, were tested using the developed fuzzy logic model. The inferences that can be drawn from such tests are
summarized below:-
By increasing the frequency of buses on the network, more commuters are found to switch over to
public transport thereby reducing private mode share in the total number of trips.
It can be seen that improvement in comfort condition of buses will result in a larger modal share of
buses only if there is no consequent increase in travel cost. In fact an increase in travel cost, even
with an increase in the comfort condition of buses, would cause a decrease in the share of public
transport. Thus the policy makers will have to find low cost solutions to increasing the comfort
conditions in the buses.