Analyzing the Impact of Traffic Volumes on Key Performance Metrics for Connected Autonomous Truck-Enabled Highways

The study investigates the impact of variable traffic volumes on connected autonomous vehicles (CAVs) net benefits. 275 scenarios were simulated. These were developed from 11 market penetration rates (MPRs) ranging from 5 to 95% for connected autonomous trucks (CAT), and 25 Average Annual Daily Traffic (AADT) from 5,000 to 250,000. For each scenario, a whole-life cost was calculated for a baseline, and then a CAT regime. The baseline regime used an existing conceptual standard three-lane motorway. The CAT regime saw remodelling the baseline to a non-standard four-lane cross-section incorporating a 2. 85m wide dedicated CAT lane and three narrow lanes for manual vehicles (MVs). There was no widening of the existing baseline carriageway to create the CAT regime: the total carriageway width remained 11m in both cases. The net benefit for each scenario was the difference between the baseline and its corresponding CAT regime costs, as monetary values. The costs included initial construction, maintenance, accidents, fuel consumption, and travel time over a 20-year design life. The study used pricing models and unit prices to determine initial construction and maintenance cost. TxME autonomous vehicles pavement analysis software and specially developed regression models were applied to determine pavement failure frequency and maintenance points for CAT and manual truck lanes. Costs of accidents and fuel consumption were calculated using historic data from UK government sources and analysing results of secondary studies on CAT safety and fuel consumption research experiments and simulation modelling. Travel time costs were determined from a complex amalgamation of inter-linked mathematical relationships involving CAT proportions, theoretical free-flow speeds, actual speeds, posted speed limits, lane widths, traffic volumes, lane capacities, lane utilisation distributions, construction cost-time models and pavement failure rates. It was found that introducing CAVs at low rates will be detrimental to network performance for many traffic flow situations (<170,000 AADT). At very high MPRs net positive benefits can be achieved from 100,000 AADT upwards