Shipping and transportation are by far the most expensive costs in logistics. In the UK and Europe only 63% of journeys carry ‘useful load’ and average vehicle utilisation is under 60%. This means lots of empty space on trucks and trains as they move around the road and rail networks.
The concept of collaborative-shipments is not new. As a solution it addresses shipping space utilisation by optimising routes through data aggregation and analysis, synchronisation of incoming orders, meeting dock windows, whilst predicting traffic conditions. It sounds relatively straightforward for a single operator, but for interdependent networks has proven too difficult to fully implement due to lack of access to the appropriate data and administrative overheads.
Co-located or geographically close suppliers do not know whether they are sending items to similar locations at similar times. For them to know, someone needs to tell them, which means they need to share their data.
For retailers, orchestrating co-loading or co-shipments involves transaction costs. The benefits of truck co-loading does not necessarily outweigh the costs of paying for manual orchestration.
For logistics providers to be the orchestrator, suppliers would need to use the same logistics providers, creating a lock-in effect.
For a third-party mediator to orchestrate, multiple suppliers need to sign up and pay for it.
Solving the challenge through shared economies
Working with the University of Cambridge, Stowga parent Value Chain Lab (VCL) initiated a study to explore the potential of shared economies and collaborative shipping for solving this challenge.
The project created an algorithmic solution to freight co-loading, so information between the actors involved in co-shipping can be shared and optimal routing solutions can be found, without the associated transaction costs.
As there are tens of thousands of independent local carriers across the UK alone, the total number of the agents needed to represent each carrier can become huge, thus a scalable, efficient and effective agent architecture is a key component to the success of handling collaborative logistics in realistic settings.
For the study, the participants templated various types of agents that define the basics of supply chain entities; and created a simplified simulated collaborative logistics environment in which experiments were run for studying the relationship between agent architectures, the number of agents and time taken to reach solutions, including collaborative vehicle routing solutions built upon this environment.
Creating a network-of-logistics-networks
While this simulated environment is at its initial stage it is a promising attempt that sheds light on agent-based collaborative logistics and we are expecting to create a more mature platform in our follow-up research.
For VCL and Stowga, this research sets the foundation for its marketplace model of creating a network of logistics networks which will defragment the industry.
Historically, logistics has been very fragmented, with manufacturers and retailers reliant on a large number of unconnected individual warehouses, 3PLs (third party logistics providers), and transport service providers.
But technological advancements now allow for the process of identifying algorithms and approaches for sharing transportation capacity and reducing empty mileage.