The future of last-mile delivery: a scenario thinking approach

`Last mile' in cities is not merely a logistics problem, but also a significant urban planning challenge. Last mile is the final leg of the supply chain that involves high-frequency, low-volume, and short-haul distribution of goods to end consumers. This last leg is the most important, yet the least efficient part of supply chain. With rapid growth of online retail transactions and increased supply chain complexity of the globalised production networks, the size, scale and scope of last mile-driven logistics problems are most likely to escalate in the immediate future. Transport infrastructure and planning controls are among the key factors that contribute to the severity of last mile delivery (LMD) problems in large cities. However, interrelationships and interdependencies between last mile logistics and transport networks and urban planning controls are neither theoretically evaluated nor empirically tested in the extant literature.

This thesis aims to measure and map the potential transportation network impedance to last mile delivery, build future last mile delivery scenarios and formulate a strategic framework to mitigate the risk of last mile delivery failure within Metropolitan Melbourne. A five-stage Scenario Thinking method was applied to understand and analyse the provision of last mile delivery and the associated `critical uncertainties'. A Geographic Information System (GIS) was used to compute and visualise the potential transportation network impedance to last mile delivery within the Metropolitan Melbourne. Spatial data representing the key constraints of transportation network and planning controls were used to compute the potential transportations network impedance to last mile delivery.

A Scenario Thinking (ST) workshop was conducted with 14 participants who represent three major stakeholders, namely operators, administrators and users. This was designed to discuss and identify key planning and transportation constraints contributing to last mile challenges and to formulate future possible and plausible last mile delivery scenarios. Thirty-four major issues that underpin last mile logistics were identified, which were clustered into six thematic dimensions through an iterative consultation process. These include: i) Freight Infrastructure; ii) Infrastructure Supply; iii) Landuse Intensity; iv) Infrastructure Sharing; v) Intersection Controls and vi) Human Behaviour. Infrastructure Supply and Landuse Intensity were found to represent higher uncertainty and higher impact on city logistics provisions. Hence, they were used to build future LMD scenarios.
Spatial data of seventeen mappable constraints, including traffic count, population density, zoning, proximity to Melbourne CBD and activity centres, intersection constraint, speed limit, number of lanes, toll, railway boom gate, traffic lights and trams routes, were standardised and aggregated using a composite index technique. The generated LMD impedance index was then mapped using an overlay function to estimate the potential hindrance to last mile delivery as imposed by built and regulatory environment. The mapped outputs illustrate significant spatial variations in LMD impedance levels across different parts of Metropolitan Melbourne. Impedance to last mile reduces with increased distance from the Central Business District, found to be high within the designated activity centres and varies across inner, middle and outer rings.

The results reveal that the future outcomes of last-mile delivery is dependent upon which scenario eventuates out of the four scenarios formulated using Infrastructure Supply and Landuse Intensity. The best-best scenario characterises an increase in infrastructure usage, decrease in road congestion, lower delivery cost, and increased last mile delivery productivity and efficiency; whilst the worst-worst scenario represents a decrease in the usage of logistics infrastructure, transport delay and congestion, increased last-mile delivery cost, loss of work productivity, reduced deliveries per day and higher environmental impacts.
Further analysis indicates the relative positioning of major stakeholders based on an interest-power matrix. Last mile logistics stakeholders such as Federal, State and Local Governments, and Transurban with high-power tend to exhibit low-interest in provisioning last-mile logistics; while those with high interest such as Drivers and Business owners have low power to influence any positive change to enhance the efficiency of last-mile delivery. Stakeholders with high-power and high-interest were identified to include Public Transport Victoria, VicRoads, Trader Association, Port Authority who could assist in policy-making to help improve last-mile delivery efficiency and curtail carbon footprint within an acceptable level.

In this study, a strategic framework is developed to support the formulation of key objectives and future-oriented actions. Five strategies along with key actions were proposed to tackle the LMD challenges associated in Melbourne. These include: land-use zoning, last mile corridor, distribution network strategies, multimodal use strategy and stakeholder engagement strategy. The land use zoning strategy can be implemented to geographically demarcate last-mile delivery zones to improve the efficiency of last-mile delivery to retail businesses within the localised area. The last mile corridor strategy would expedite last-mile delivery along the main arterial networks through the development of linear freight routes to improve last-mile connectivity between key business hubs. This geo-targeted strategy will help reduce the environmental footprint of last-mile delivery, ease traffic bottlenecks and potential conflict between last-mile delivery trucks and commuters by confining truck movements to designated routes. The distribution network strategy would promote consolidation of goods through a holistic integration of people, facilities and transportation infrastructure as a single unified city logistics network to support the development of Urban Distribution and Consolidation Centres in vicinity to the Activity Centres. The Multi modal use promotes LMD using trucks and train, trams or bicycles as an integrated system. The stakeholder engagement strategy places a greater emphasis on the shift in interest of stakeholders with power to cause positive change through advocating for the implementation of effective policies and regulations.

Overall, this thesis is the first study that applied the Scenario Thinking method along with GIS to construct, measure and map the potential last-mile delivery impedance. Scenarios were formulated to provide improved understanding of future last-mile delivery. Strategies were recommended to help develop operational plans and deploy future investment to tackle the challenges associated with the worst/worst scenario. The mapped outputs improve the future understanding of the complex interactions between transportation infrastructure, planning controls and last-mile delivery. This understanding and proposed strategies in turn would help enhance the efficiency of last-mile delivery in the context of large cities.