ITS Monday: Edition 17, 2025

MaaSformer-MMOE: Multi-task Transformer under mixture-of-experts framework for MaaS bundle customization

First up this week, a new paper from academics at the Institute of Transport and Logistics Studies, with co-authors Haoning Xi, Zhiqi Shao, David Hensher, John Nelson, Huaming Chen, and Kasun Wijayaratna. The abstract:

Mobility-as-a-Service (MaaS) platforms are reshaping urban mobility by integrating multiple travel modes into seamless, user-centric systems. However, designing dynamic MaaS bundles that adapt to user-specific preferences, evolving over time in response to changing travel behaviors and shifting needs, remains a significant challenge.

The rise of big data and artificial intelligence (AI) has unlocked new opportunities for data-driven personalized MaaS bundle solutions. In this study, we introduce an innovative MaaSformer-MMoE framework to customize user-specific monthly MaaS bundles by predicting each user’s mode-specific usage frequency class (classification tasks) and travel fare (regression task) for the upcoming month based on the user’s previous travel records.

Within the multi-gate mixture-of-expert (MMoE) framework, each expert network is a MaaSformer, and each gate determines the weighted contributions of expert outputs relevant to a specific task tower. MaaSformer integrates two key modules: 1) Multi-mode Transformer processes continuous time-series features (e.g., monthly travel time, distance, and fare) employing a multi-feature self-attention mechanism; 2) OD Transformer processes origin-destination (OD)-specific travel features (i.e., journey frequency) using a multi-OD self-attention mechanism.

Evaluated on a multimodal (i.e., bus, rail, ferry, and tram) dataset of over 1.5822 million users in Queensland, Australia, from 01/2021 to 01/2023, the proposed MaaSformer-MMoE demonstrates state-of-the-art performance in predicting mode usage frequency class and travel fare compared with 9 baseline models, significantly improving user satisfaction, adoption and retention for MaaS platforms.

Related iMOVE articles:

• Working from Home: Info, Projects & Resources
• Prospects for Working from Home: Assessing the evidence
• Traffic Congestion: Info, Projects & Resources

Drivers of SUVs and pick-ups should pay more to be on our roads. Here’s how to make the system fairer

A piece in The Conversation, by Milad Haghani, Associate Professor & Principal Fellow in Urban Risk & Resilience, The University of Melbourne.

“SUVs and light commercial vehicles comprise almost 80% of the market. Four in five new vehicles sold in Australia today are an SUV, ute, van or light truck. As larger vehicles become the new norm, they bring more road wear, urban congestion and demands on infrastructure such as parking. It’s time to ask: should drivers of larger vehicles pay for the damage and disruption they cause, through higher registration charges? Generally, yes. Bigger cars mean bigger costs for everyone else. It’s only fair those costs are reflected in how we price their use of public roads.”

NSW to legalise e-scooters

An 8-minute audio interview with Milad Haghani (his second inclusion in this week’s ITS Monday) on the Radio National Breakfast show. “In NSW, it’s legal to buy an e-scooter but not legal to ride one. Now the NSW government wants to regulate this popular form of transport.”

Related iMOVE articles:

• Active Transport: Info, Projects & Resources
• Micromobility: Info, Projects & Resources
• Road Safety: Info, Projects & Resources

Drivers’ trip purpose and pre-crash maneuvers as determinants of pedestrian injury severity: Insights and policy recommendations

A new academic paper co-authored by Matin Sadeghi, Kayvan Aghabayk, MohammadAli Seyfi, and Nirajan Shiwakoti. The abstract:

Pedestrian safety is a critical component of sustainable urban mobility. Despite the known societal and health benefits of walking, pedestrians remain one of the most vulnerable groups on the road, often suffering severe injuries in crashes due to their lack of physical protection. To address this issue, policies encouraging walking must go hand in hand with measures that enhance safety. This study investigates the risk factors affecting pedestrian crash severity on urban roads in the United Kingdom, with a focus on the underexplored role of drivers’ trip purposes and crash preceding maneuvers.

Using a temporal analysis, the study reveals that drivers’ trip purposes, particularly ‘commuting trips’ and ‘work-related journeys’, are significantly associated with serious pedestrian injuries. The analysis further explores crash preceding actions, finding that vehicle maneuvers such as reversing, turning, and stopping/starting, as well as pedestrian movements such as crossing from the driver’s offside or walking along roadways, significantly impact injury outcomes. While minor injuries are more likely during vehicle turning, reversing, and pedestrian road-walking, fatal injuries are strongly associated with maneuvers like pedestrian crossings that involve higher risk.

These findings provide essential insights into pedestrian crash dynamics and can inform more targeted safety measures. Policy recommendations include developing tailored traffic management strategies, such as limiting certain high-risk maneuvers in pedestrian-dense areas, lowering speed limits in urban areas, and adopting advanced vehicle safety technologies. Additionally, creating awareness campaigns that emphasise the dangers posed by specific driver behaviours during work-related or commuting trips could help mitigate risks. Enhancing driver training for taxi, truck, and bus operators, particularly in urban areas, and implementing workplace safety programs targeting driver fatigue could improve road safety.

Finally, public awareness campaigns emphasising both pedestrian visibility and safe crossing behaviours are crucial for reducing pedestrian vulnerability. Integrating these insights into broader urban transport planning and adopting stricter pedestrian safety regulations would support more effective interventions aimed at reducing pedestrian injury severity and enhancing road safety in urban environments.