Deliverables

To create a recommendation for relay station locations, the team used historical data from Coyote’s six months of shipments from March 7, 2022 until September 7, 2022. The six months of historic shipping data was used to determine an initial area for examination due to the large scale of Coyote’s operations in the United States. Delivering a recommendation for locations throughout their entire network would have been impossible; therefore, the team sought to provide these locations along a single Coyote lane as a proof of concept, allowing Coyote to test the results on a small scale before expanding the system nationwide.

It became apparent to the team during initial research that there were many factors that affected time on the road for truck drivers that could not be extracted from the data: high traffic volume, poor weather conditions, and trucker’s preference in route navigation are among the most unpredictable. Furthermore, the route’s status as a proof of concept required it to be useful towards Coyote’s future expansion plans. Therefore, the team decided it would be in Coyote’s best interest to provide them with multiple routes that the company can examine using factors that the team wasn’t given access to, as well as factors that may simply not have presented themselves yet for the future of the relay station system.

As one of the options presented to the client, the team decided to create a route using only Coyote’s most simple constraints; this became known as the “naive route,” referring to the naivety of the assumption that no optimization past the initial constraints was required that led to its creation. Since the majority of Coyote lanes are located north of Texas, a more northern route should also theoretically be more beneficial for future expansion of the system. Using these constraints the team created a route stretching between Greer and Laredo containing relay stations in urban areas, per Coyote’s request, in order to ensure that there is a large enough population that there will be enough drivers to staff these routes.

To ensure that the relay stations chosen have high potential to accommodate other routes in the future as well, the team used a Traveling Salesman Problem (TSP) approach to evaluate the route between Greer and Laredo. The objective is to select a set of relay station cities with high frequency of loads in and out, while minimizing the total distance traveled between Greer and Laredo.

It is important to not increase the delivery time to deliver from Laredo, TX to Greer, SC or Greer, SC to Laredo, TX when choosing the cities with high potential. According to the client’s data, the average distance traveled between Laredo and Greer by a driver was 1302 miles. Considering Key Assumption 1, it can be assumed that a shipment traveling on this route would usually take 3 days. If the relay station network's total distance traveled results in the shipment taking longer than 3 days the shipper would experience a delay. 

The objective of the model uses a weighted sum method by assigning weights in terms of importance on the two sub-objectives: (1) maximize the total sum of the number of routes the relay network accommodates within the six-months data and (2) minimize the total distance traveled between Laredo and Greer.

The team’s second deliverable was a financial analysis in Excel that will allow Coyote to input costs per mile, hiring cost, cost per relay station, number of relay stations and drivers, number of shipments, driver retention rate, annual deadhead, profit margins, and revenue split, to estimate the monetary value of each route to Coyote, the carriers, and the drivers. Coyote will be able to use this document to analyze the relay stations’ costs, benefits and to decide their break-even point.