Document Type

Conference Paper

Publication Date


Subject Area

mode - rail, place - australasia, planning - travel demand management, technology - ticketing systems, operations - capacity, operations - crowding


Melbourne, peak congestion, investment, rolling stock, non-peak, morning peak, afternoon peak


Melbourne, like many large cities around the world, experiences significant peak congestion on its public transport network. Demand for services in peak times is a key driver of investment in new rolling stock and infrastructure. A way of delaying these significant investments is to better utilize current available resources in non-peak times, by spreading out peak demand over a wider time period.
This study used an online survey methodology to investigate the propensity of peak period train passengers commuting for work on Melbourne‟s Pakenham line to shift out of peak travel times to access a better price, service frequency, stopping pattern and train conditions. The main methodology was two Discrete Choice Model exercises (one morning-peak and one afternoon-peak) which systematically varied time of travel, price and service attributes to model customer behaviour under various scenarios.
The results support the view that demand can be influenced by price and service attributes, and support the development of detailed business cases for reducing peak demand. Implications are discussed, as well as the challenges in converting these customer behaviour predictions into workable timetables, and in accurately costing the benefits of delays in investments in new rolling stock and infrastructure.