COO – “How are we going to modify our transportation network to account for this sudden spike in diesel prices?”
VP, Transportation – “We will run some scenarios and get back to you in a week or two.”
COO – “Excuse me?”
This is a conversation no transportation executive (or any executive for that matter) wants to have with their leadership. Companies always need to be in pursuit of continuous improvement opportunities and risk assessments akin to the fuel problem posed above. The classic analytical process involves companies gathering and cleaning historical data, setting up a transportation modeling exercise deploying an appropriate software application and running scenarios to determine which if any should be deployed. The process to set up and run these scenarios is often very time-consuming.
The implementation of a transportation digital twin offers a better approach. For companies who operate a transportation management system (TMS), set up an identical version of the constraints and routing guides used to convert transportation orders into shipments. This can be done deploying a number of transportation modeling applications available in the marketplace.
Congratulations – you now have your baseline model established against which your scenarios will be compared. Using a baseline order set, and more importantly, the orders that feed the TMS going forward, there are myriad scenarios that one can test:
- Mode shifts
- Multi-stop truckload opportunities vs. LTL?
- Truckload vs. Intermodal?
- Service expansion or contraction (i.e. 2-day delivery vs. 1-day delivery)
- What is the cost tradeoff for expanding delivery times?
- What is the cost to the business for offering narrower delivery times?
- What about expansion of unattended or weekend delivery? Or both?
- Optimizing deployment of a dedicated or private fleet
- How many power (tractor) and driver resources should be in the fleet?
- How is fleet sizing impacted based on the cost of fuel?
The common denominator of deploying a transportation digital twin is enhanced decision-making in a much more compressed timeframe than classic modeling exercises. Real-time data feeds into the model, allowing planners and operators to test scenarios and predict outcomes before implementing changes in the physical world. This reduces risks, minimizes disruptions, and helps allocate resources efficiently.
—Brian Fish, St. Onge Company
