University of Southampton
Effects of connected autonomous trucks on highway infrastructure performance: a whole-life costing approach to design optimisation
Abstract
dc:description.abstractConnected autonomous trucks (CATs) offer the prospect of vehicles executing highly defined, repeatable trajectories. This affects lane widths and pavement requirements, which must be assessed jointly as they are interlinked. Applying current design standards could lead to suboptimal designs because of conservatism. An optimised highway design model is required to ensure that autonomous vehicles are not detrimental to the network.<br/>This study investigated the costs and benefits associated with innovative non-standard road cross-section remodelling options to accommodate CAT. Fifty cross-sections were generated, operating under 11 CAT ratios, resulting in 550 unique scenarios.<br/>Over a 20-year design, this study used pricing models and unit prices to determine initial conversion and maintenance costs. TxME autonomous vehicles pavement analysis software was used to calculate the pavement failure frequency for CAT. Specially developed regression models were applied to calculate the same pavement behaviour parameters for manual trucks.<br/>Accident, carbon dioxide emissions (CO2e), and fuel consumption rates for manual vehicles were derived from historical data compiled by the UK government. CAT rates were calculated by analysing the results of the secondary studies from simulation modelling research. The overall travel times were determined from a complex amalgamation of inter-related mathematical formulae involving CAT proportions, theoretical free-flow speeds, actual speeds, effective speeds, posted speed limits, lane widths, traffic volumes, lane capacities, lane utilisation distributions, construction (conversion) cost-time models, and pavement failure rates.<br/>The optimal highway design for low to mid-range MPR values was non-standard four-lane cross-section with 2.85m CAT lane, 3m manual truck lane, and two 2.575m passenger car lanes. At higher MPR values the optimal design is a standard three lane section of widths 3.65m, 3.70m and 3.65m. Demand-based sensitivity analysis of a further 275 scenarios found low CAT rates are detrimental to performance metrics, i.e., where daily vehicular flows are under 170,000. Beyond this demand, net benefits rise exponentially.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D.
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Southampton
- Year dc:date.issued
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Jehanfo, Hameed
- Advisors dc:contributor.advisor
-
- Preston, John
- Kaparias, Ioannis
- ORCID person:identifier.orcid
Rights
- Language dc:language
- English, eng