Less-than-Truckload freight planning problem: Designing the service network for fleet automation

Al Hajj Hassan, Lama

doi:https://doi.org/10.21985/n2-e4hk-3s12

Work

Less-than-Truckload freight planning problem: Designing the service network for fleet automation

Public

Download PDF

Download All Files (.zip)

Autonomous trucks (ATs) are envisioned to potentially revolutionize the logistics sector. The integration of autonomous trucks into the overall truck fleet will impact industry regulations and ease driver-related challenges. It also has the potential to improve road safety and reduce carrier costs. There is an extensive body of literature on long-haul consolidated freight service network design (SND) with manual-driven trucks. However, there is limited literature that studies network design with automation and none comparing different levels of automation. This thesis develops and tests formulations and solution methods for freight service network design that account for different formats of autonomous truck deployment. The goal is to help service providers manage fleet operations and to present insights to guide policy makers in preparing for autonomous truck deployment. To meet these objectives, the thesis first revisits the taxonomy of the freight planning problem, identifies changes needed to existing categories, and suggests additional modeling challenges that should be examined at each level of autonomous truck deployment. Next, the thesis examines the impact of driver work rules and regulations on network planning. The results show that the driver return to domicile requirement has a more significant contribution to cost than the hour of service regulations. Under the current fully manual fleet situation, the driver cost is the main determinant of routing choice compared to the other cost components (vehicle, fuel and handling costs). The thesis develops and tests modified service network design formulations that account for five levels of truck automation: (1) manual operations (base case), (2) mixed fleet with drivers on board each AT, (3) mixed fleet with two types of ATs (with and without a driver on board), (4) mixed fleet without a driver on-board AT, and (5) fully autonomous fleet. The computational experiments show that the cost savings with AT deployment result from partial or complete elimination of driver costs, reduction in empty miles traveled and the decrease in the number of trucks required to service the loads. Deployment of ATs is preferred over the long-haul direct trips connecting terminals. AT deployment even when restricted to specific geographic zones has operational benefits and environmental advantages driven by the reduction in costs and decrease in unproductive miles driven. The results show that the network routing may not exhibit significant change with AT deployment. This is beneficial for carriers as that would mean their current terminal locations and capacity may not be impacted and may not have to be adjusted (relocation, increase capacity) with deployment, assuming the demand patterns remain unchanged. Finally, the thesis examines daily load planning in LTL operations for the five deployment scenarios defined in Chapter 5. Given daily updates of load information, the paths for the five deployment scenarios were adjusted using two daily updating methods. Both methods start with a base plan in which paths are generated based on the historic daily distribution of load dispatches during an average week .The two methods are: (1) Option 1: re-optimization of pre-booked loads and new requests, and (2) Option 2: optimization of new requests only. The solutions of the two options are compared to the hindsight plan which assumes complete information of actual requests placed. The results show that the cost savings achieved with re-optimization compared to insertion increase with more demand variability; this outcome is consistent across all fleet mixes. The percentage of empty miles driven with re-optimization is close to the “Hindsight” plan. When the majority of the loads are new arrivals the reduction in computational time achieved by insertion is more costly than when the majority of loads are pre-booked. Therefore, complete re-optimization is favored to insertion (as implemented here). Nevertheless, if solution time is an important factor, then it becomes critical to develop efficient algorithms for this purpose. With daily re-optimization, the majority of the plan changes adjust the terminals visited by the load compared to just changing the dispatch and arrival times along the load’s path. The distribution of plan changes by type with re-optimization are similar across all fleet mixes suggesting that the main determinant for path changes is the load information availability and not truck type. Summarizing, the main takeaways for LTL carriers: • AT deployment is expected to reduce cost of operation, percentage of sleeper teams deployed, empty miles driven, and fleet size requirements. • AT deployment even when restricted to specific geographic zones has operational benefits. • Network routing may not exhibit significant change therefore current infrastructure (terminals) is sufficient assuming current demand patterns remain unchanged. • Daily re-optimization of load plans can reduce costs and empty miles travelled in all fleet types. • Daily re-optimization mainly changes the terminals visited by the load however the majority of the loads follow their original plans. The main conclusions for policy makers show that there is a need to: • Support testing AT deployment over long-haul interstates. • Maintain adequate infrastructure along AT corridors. • Examine potential locations for autonomous truck ports that act as transfer hubs between long-haul ATs and manual short haul trips. • Establish uniform policies across different states and revisit hour of service rules for each deployment level. • Prepare guidelines for proper interaction between autonomous trucks and passenger vehicles on the road.

Creator