An Approach to Measure the Impact of Emerging Technologies on Urban Freight Productivity 

Abstract
While freight trucks represent a small portion of traffic volume in urban areas, they are a critical contributor for energy consumed in the transportation sector. To respond to significantly increased freight activity, especially in urban areas and improve delivery efficiency, new delivery vehicles are under development, such as electric trucks/vans, drones, and delivery robots. Understanding the impact of such vehicles on urban freight activity is critical to capturing the freight productivity (or lack thereof). While modeling and forecasting freight travel and energy impacts are the foundational step to achieve this, placing a metric around fright transportation and energy efficiencies is what really helps decision makers compare and contrast different technologies and business models. However, conversational metrics such as truck-miles, ton-miles, or value-miles evaluate freight performance at the aggregate level, and they are often unidimensional. They have critical shortcomings that limit them from accurately capturing the impact of emerging technologies on the productivity and efficiency of freight systems with little emphasis on the energy intensity of transport mode.

In order to overcome such limitations, this study proposes a novel and practical methodology for development of a comprehensive freight mobility and energy productivity metric, labeled the Freight Mobility Energy Productivity (F-MEP) by integrating accessibility theory that has been applied in passenger and freight travel behavior studies (1–6) with vehicle operation time for urban delivery and energy efficiency of mode. The F-MEP is defined as a metric that can measure how effectively freight can reach its potential locations (i.e., end customers) from a given location through various modes including emerging vehicles. The new metric is designed to incorporate characteristics of freight movement in a city such as high sensitivity to congestion, high circuity and different tour types in a single formulation. The methodology developed in this study is implemented for Columbus, OH using a hexagonal grid representing each location for metric calculation. Five delivery vehicles are considered: three conventional delivery trucks, delivery robots, and e-bike. Vehicle operational characteristics have been identified from literature and used as input variables for the F-MEP calculations. Five primary location (opportunity) types associated freight activities are identified based on land use data provided by CostarTM including the establishment (representing business deliveries) and population (representing residential deliveries). Results show intuitive trends, both at the aggregate level (i.e., overall freight productivity) as well as at the disaggregated level of mode-tour type F-MEP calculation. Disaggregated results demonstrate how different the freight performance metrics (reflected by the F-MEP score) are among different modes. This capability is an extremely useful feature of the F-MEP metric that enables users to conduct customized analyses to understand the impact of different emerging technologies on freight performance depending on the locations. The novelty of this study lies in the first attempt (to the knowledge of the authors) to develop a unified freight performance metric by considering multidimensional factors (time, energy, and cost) of urban freight system.