Winter Operations Uses Route Optimization to Improve Road Safety

Throughout Colorado, there are 224 different patrols in charge of a clearing, maintaining, and monitoring the condition of the roads during inclement weather. Segments of each road within a patrol’s assigned region are distributed to individual Colorado Department of Transportation (CDOT) workers who are then responsible for keeping their segment safe for public use. They do so by determining the treatments and amounts of plowing their section needs, and then consistently drive the roads themselves to monitor any changing conditions. All of this work is managed and executed by CDOT’s department of Winter Operations. 

Unfortunately, some of the designated routes of these patrols have not been updated in the past 30-40 years. Manager of Winter Operations, David Johnson, and Director and Director of Highway Maintenance, Kyle Lester, decided it was time to re-evaluate the efficiency of these routes. To do so, they have begun a ‘Dynamic Route Optimization’ program. This program will utilize innovative technology to compute data on a variety of different climate and roadway factors to then produce live maps of the conditions of every road segment in the State. The data will then be used to evaluate the efficiency of existing patrol routes. 

How the Technology Works

This innovative Route Optimization Program was the brainchild of both Johnson and Lester, but in order to turn their idea into reality, they needed an outside party to create the actual software. Johnson and Lester decided to use Vaisala as their sole source vendor, due to their past cooperation with CDOT  in the development and utilization of friction sensors, and their already vast knowledge about Colorado's infrastructure, as they are based in the state. 

Currently, the framework for the program is being laid out by first creating a static optimization software. Essentially, a static optimization software does route optimization based on unchanging variables such as identified problem areas and topography. Once the static optimization is fully mapped, then they will begin to input live data, such as weather, traffic conditions, the risk of snow drifts, and snowfall amounts. 

“I see there being a point where all the friction data we are collecting will actually be put into how we optimize routes,” Johnson added, “I think that would help justify and validate some of the problem areas people are saying are there, but we haven't really proven through data yet”. 

For the final product, they are working to develop the program so that superintendents and deputies can go into the adaptive platform and set parameters based on any incoming storm, then send that mapped data to their employees. All drivers would have access to these maps, but would not be able to adjust the controls. 

For example, “if there is a region that the supervisors predict will have a lot of blowing snow and avalanches, then they can adjust the weighting of the variables to mark areas as more important so that the software will then identify which areas should be a top priority for the patrols” explained Johnson. The drivers would then be alerted of their updated route. 

In order for this technology to be truly effective, supervisors will need to be well trained in navigating the interface and the interface must also be very simple to customize. During big storms, the supervisors need to focus on strategizing how to make the roads safe as quickly as possible. Johnson and Lester plan to develop the interface so that supervisors will have access to a menu panel, with switches to turn on-and-off different attributes, or a drop-down list to select the category of attributes they want to look at. 

How This Affects Drivers

The most obvious way that this program will affect the general public is in the time it takes for an individual to get back on the roads and safely get to where they need to go. The number one concern at CDOT will always be to keep roadway users safe. 

“The main purpose of this project is increasing safety and saving lives through improved cycle times,” Johnson explained. 

There are also cost savings associated with the program being implemented. “I would estimate that 60-70% of routes could become more efficient," Johnson claimed. The Winter Operations department is constantly working to save CDOT funds. In fact, last year, while the department was only beginning to test this route optimization program, the department as a whole saved $17 million compared to the previous year. This is in part due to last year being a weaker than usual winter for Colorado, but the savings do still show an effort being made by the department. 

“While it is difficult to quantify savings, you can definitely see results through reduced crash patterns, communications from the public, and reduced cycle times,” explained Johnson, “Those are the types of results I really expect to see from the implementation of this technology.” 

Leading the Way for DOTs

“No one in the US is currently doing dynamic route optimization like Colorado,” said Johnson. While other DOTs have implemented optimization programs, they are currently using ‘out of the box’ software. UPS and other delivery companies use similar programs, as well2. However, that type of software only computes static data and is only really beneficial to use within cities. 

This program has incredible potential to be implemented across the country in all DOTs. While CDOT’s focus has been to create it for snow plow optimization, this same program could be used to assist in a number of natural disaster situations. 

While this project has made great progress, it is still in the development phase. “There is still quite a bit of back work that needs to be done to make this project vision attainable,” Johnson explained. He also expressed that while this project does not currently have a deadline for complete roll-out, it will definitely not be finished by next winter.