Digital Twin

Every day, our frontline operations team has to switch between systems to manage the flow of the Southern California Drayage business. There are 13 different systems – across terminals, transportation management systems (TMS), chassis providers and yard management systems (YMS) – that they may need to consult to manually plan operations and dispatch drivers.

This multi-system dance makes operations complex and inefficient, increasing the risk of error and delay. In some cases, this can lead to unnecessary fees such as demurrage, detention and chassis misuse fees.

These existing systems provide milestone updates, rather than a continuous picture of operations. This makes it difficult for teams to work together, as each business line only receives updates via their own system.

This lack of a central visibility leads to siloed operations and difficulties, such as the inability to run “what if?” scenarios. Teams struggle to plan for any future developments or changes in the operating model.

We started the project by weaving together our existing systems into one centralized management platform where dispatchers, customer service representatives, asset managers and management can all work off the same pane of glass.

To integrate these systems we used methods including direct API integrations and RPA-assisted integration depending on the ability of our source systems.

From there, we combined existing data with additional data sources, including real-time information (IOT devices, ELDs and TMS) and third-party sources (traffic patterns for heavy vehicles), to develop an optimizer that helps our transportation teams design the best sequences for processing inbound containers.

We also incorporated the management of a new fleet of electric trucks; these vehicles require more downtime and have a more limited range than a traditional diesel (ICE) vehicle. By adding this variable into the digital twin’s algorithm, we are able to plan charging downtime and improve vehicle utilization.

Putting all this data together creates a complex system of constraints, including labor and asset availability, customer priorities and unplanned events such as non-workable loads and late deliveries from the terminal.

With this information, the tool creates an ideal plan for the day, assigning drivers and vehicles and determining the best order for drop offs and pickups. The data generated by the tool also allows us to target transactions where something has gone awry, either in the technology or the physical environment.