Logistics Network Analysis Model of E-Grocery Built with a Simulation Tool

Over the last two decades, home delivery of grocery items has emerged as a comprehensive alternative to stationary grocery shopping and can effectively aid in reducing traffic emissions.

In 2017, 17.5 % of the entire traffic volume from motorized private traffic in Germany originated from grocery shopping trips. Therefore, by promoting the utilization and growth of e-grocery and consequently reducing or avoiding private errands and shopping trips, traffic loads and traffic-related emissions can potentially be decreased.

To assess the ecological value of e-grocery compared to stationary grocery shopping, the researchers proposed a simulation approach to model both mileage and emission outputs.

The model for logistics analysis was built with AnyLogic (version 8.5.2) simulation tool and combined agent-based simulation (ABS) properties with discrete-event simulation (DES).

ABS can effectively be used to model environmental uncertainties due to the high amount of autonomous and heterogeneous components and dynamic interdependencies between the system units. It also allows for studying both structural and functional aspects of complex networks (such as grocery logistics networks) and alter simulation properties for experimentation purposes.

The logistics network simulation model was built on an industry partner’s data on delivery operations from the central warehouse to the final customer.

With each simulation run representing one particular day, the evaluation of the results was based on e-grocery utilization rates. Those rates showed the potential benefits or drawbacks of grocery home deliveries in terms of traffic influences and emission outputs.

Applied to the operational case of a major retail organization in Hanover, Germany, the logistics network analysis performed with the simulation tool showed that a high e-grocery utilization rate can aid in decreasing total driving distances by up to 255 % relative to the optimal value as well as CO2 emissions by up to 50 %.

Logistics analysis simulation-based tool

Research approach for the simulation study