Effects of pedestrian–robot interaction on delay and CO2 emissions in last-mile delivery

This study examines how minimising human–robot interaction influences travel delay and carbon dioxide (CO2) emissions in sustainable last-mile delivery. An analytical framework is proposed to integrate pedestrian density, robot travel delay, energy consumption, and route-level carbon dioxide emissions within sidewalk environments. Pedestrian density is calculated at the link level and analysed using a linear mixed-effects model with a link-specific random intercept to account for repeated observations within sidewalk links. Field experiments using an outdoor autonomous mobile robot are conducted to obtain travel-time observations and estimate the delay relationship. Energy consumption is estimated based on the adopted average operating power and recorded travel duration, and route-level carbon dioxide emissions are calculated using electricity emission factors. The results indicate that congestion-prone areas, particularly near intersections and transit stops, are associated with increased travel delay and higher estimated operational emissions. These findings suggest that minimising human–robot interaction through route planning is an important factor in improving the operational efficiency of last-mile delivery robots.