Replacement bus patronage counting and wait time measurement
This project aims to identify and trial suitable system(s) for automatic real-time patronage counting for replacement bus services during rail disruptions in Melbourne. It will focus on passenger detection technologies that can be applied in passenger waiting areas rather than installed on buses.
It will include a field trial to test two potential solutions which include video detection and the use of sensor mats. The field trial is intended to determine the advantages and limitations of each solution. This will be followed by comprehensive analysis of results and recommendations of a scalable system for wider implementation, at reasonable costs.
Project background
Rail services in Melbourne are at times replaced by bus services due to scheduled track maintenance or upgrades (planned disruptions), but also due to network incidents (unplanned disruptions). Historically replacement services were required infrequently and therefore there has been limited automation and technological development in the planning of these services. The current, and planned, high level of infrastructure upgrades means that they are now much more frequent and impactful. Additionally, the resource requirements and cost of replacement services for such disruptions is significant.
The Level Crossing Removal Project (LXRP) and its alliance partner Metro Train Melbourne (MTM) are responsible for replacement bus services, and desire to provide these as efficiently and with minimal impact on commuters and road users.
Currently, the planning of the replacement bus allocations relies primarily on predicting travel demand based on Myki data and partial records from previous disruptions. While this data is useful, the lack of full and accurate data on replacement bus patronage limits LXRP and its alliance partners to more accurately plan for actual travel demand. This current lack of data also means there is a sub-optimal feedback loop through which an evidence base can be used to estimate both current performance and future service level needs.
This means that there is currently a risk of:
- Over-procurement of replacements buses (which are then under-utilised by customers);
- Under-procurement of replacement services, resulting in a negative customer experience, through crowding or long wait times.
During a disruption, the responsibility of determining patronage numbers falls on customer service staff who may have to multi-task. Without reliable and extensive real-time information on patronage on replacement buses, there is a real challenge for the control centre to respond quickly to the changing real demand in the field.
It is also anticipated that if a successful alternative technology can be identified and trialled, this may present other value-add options or opportunities for consideration by the Department of Transport or LXRP. These opportunities would potentially relate to the creation of a new data set of travel behaviours and insights into a market segment (i.e. disrupted commuters) that currently remains hidden from existing data sources.
There may be service integration opportunities which may arise from potentially having a real-time feedback of customer patronage of replacement bus services and tracking replacement bus services. This could allow more tailored services to respond promptly to unanticipated peaks and troughs in patronage on replacement services, improving both the efficiency of service provision and responsiveness to customers.
Project objectives
The project aims to address the data needs associated with bus replacement services for train disruptions to evaluate utilisation of replacement buses and provide the basis for enhancing these services and customer satisfaction in the future.
The main aim of this project is to identify and trial suitable systems for automatic real-time patronage counting for replacement bus services during rail disruptions. The trial is intended to test multiple potential solutions and determine the strengths and limitations of each solution. The test systems should be trialled on a basis that they can be scalable for wider implementation at reasonable costs.
The primary objectives of the project are to:
- Analyse the technological options available to LXRP and DoT to address the current data gap in the performance and patronage of replacement bus services.
- Identify a preferred technology option or options to be trialled.
- Undertake and assess the trial.
This is a really interesting project. I am currently working for MTM on the Metrotunnel Project on using Video Analytics to undertake crowding analysis. I would be really interested in hearing more about how various other technologies might be used to approach this task and what sort of results were achieved.
Agree on opportunity for substantial improvement in replacement service provision. Have experienced both failure modes described (inadequate resourcing with long waits and overcrowding, and empty replacement buses) and it appears getting resourcing right is difficult, particularly when at short notice. As a bicycle rider who takes bikes on trains from time to time, replacement buses are seriously useless.
I’ll be very interested to find out which technologies you end up trialling and the pros and cons of each. So many options on the market already but linking them into what works here is not always straightforward.