Although Strategic Movements may be used as a manual dispatching tool, or to maintain and dispatch master schedules of static routes, it provides the greatest savings when used for dynamic routing.
Dynamic routing, also known as automated routing or autonomous routing is the technique where the work of a human route planner, load planner or trip planner is done by a computer process. This process must consider all these issues:
- Work location: The actual physical location where the work is to be done. This is more than a using a town name or ZIP code. It includes an understanding of how the location is accessed using the company’s vehicle. Just because two pins on a map appear to be near each other, there may not be a direct usable road to drive between them.
- Physical constraints: Each vehicle’s cube and weight limits, and equipment characteristics like the vehicle having a lift gate or a side door.
- Operational constraints: The worker’s daily start and end times, and the list of allowable tasks for a worker, based on training, business practice or regulation.
- Financial constraints: Using the least costly vehicle for the work to be done, and reducing overtime by assigning work to another worker, as needed.
- Service commitments: Promised delivery times.
All of these can be done, and have been done, by the human planning the day’s work, but there are two questions that need to be answered now:
How long does it take to plan the routes? And, how efficient is the overall solution when considering all the routes planned?
The classic way to manually assign work is to print delivery tickets, group them by location and then assign tickets to individual routes. After this has been done, the routes may need further adjustment to ensure that all issues listed above are considered.
How large a problem that can be done efficiently using a manual process depends on the skill of the planner. Every planner can visualize a solution with some number of stops, but at some point the planner has to rely on a divide and conquer strategy, where the large problem is broken into multiple smaller problems, each one to be solved independently.
With all the sorting and shuffling of delivery tickets required, and with side notes for vehicle capacities and other constraints, it does take time to do this assignment process well.
And when the process is done, the results are never as efficient as an artificial intelligence process performed by a computer.
The computer finds better assignment solutions by not using a divide and conquer strategy. It has no biases and is not influenced by how the routes were previously done. It considers all the issues at one time and learns by trying different combinations of possible solutions, working toward finding a few very efficient solutions.
Interested? Contact us at email@example.com
Like the blog? Sign up and be the first to hear what we have to say.