Traffic signals play a significant role in improving travel times and reducing traffic congestion. One novel adaptive traffic signal control technology advances from typical actuated traffic signal timing operations toward a system that responds to dynamic traffic conditions in real time. Recent research made significant progress toward testing this innovative traffic signal control technology in the field.

Max-pressure traffic signal control is a machine learning algorithm that controls traffic signal timing changes based on the number of vehicles approaching or waiting at an intersection. In this phase of an ongoing research project, investigators demonstrated the max-pressure system is compatible with actual Hennepin County traffic signal control hardware. Simulations of the max-pressure signal controller using the geometry and traffic data of a real county intersection showed the traffic signal responded to real-time traffic volumes, maximized the vehicles allowed through a green light and minimized driver delay. Next, the research team aims to test the max-pressure traffic signal control system in the field to validate its safety, functionality and operation at intersections.

Countermeasures designed to reduce pedestrian risk on Native American reservations in Minnesota successfully increased safety. However, monitoring at the countermeasure locations found that pedestrians often did not properly use the countermeasure, making it less effective. Future countermeasure design and implementation efforts will continue to use guidance stemming from a collaboration between MnDOT and the Anishinaabe Nations to improve results.

This project monitored 23 pedestrian crossing locations with safety concerns. The monitoring results led to the installation of countermeasures at six locations across three reservations. Observations at the countermeasure sites revealed an increased level of safety, but also regular misuse of the countermeasure. For example, pedestrians actuated a flashing crosswalk beacon intended to alert drivers less than half of the time.

Upgraded inductive loop technology could provide valuable vehicle count and classification information from an expanded list of locations. This innovative technology leverages the use of existing inductive loops and is 95% accurate at classifying vehicle types after hardware and software upgrades.

MnDOT upgraded existing inductive loops at five sites to evaluate vehicle classification accuracy. While the average accuracy was 95%, the technology was not as accurate for classifying trucks, which have similar undercarriage characteristics. However, reducing the number of vehicle classes from the Federal Highway Administration’s 13 classes to the Highway Performance Monitoring System’s seven classes resulted in a 97% accuracy in correctly classifying single trailer trucks. Additionally, the upgraded inductive loop outperformed another commercial vehicle classification system in vehicle classification (92% vs. 86%) and detection rate (100% vs. 77%).

Stormwater runoff on Minnesota’s roads carries debris and pollutants from vehicles. Sufficiently vegetated roadsides can catch and filter stormwater before it reaches streams, lakes and communities. But to thrive, this vegetation needs organically rich, permeable soil, which is not always present after road construction. Amending roadside soil with locally available industrial by-products is a potential solution for creating sustainable soil.

In this study, researchers collected and tested nine materials, ranging from sawdust to recycled concrete aggregate and beet tailings, to determine their potential for creating engineered soil mixes. Biofiltration performance, pollutant levels and adsorption rates, and plant growth assessments illustrated the materials’ capabilities to support robust, vegetated roadsides. Dredge sand, coarse street sweepings and ash sawdust exhibited significant potential as roadside soil amendments. Life cycle analyses and a design guide will help road engineers choose materials for sustainable, effective soil mixes for roadside stormwater management. 

Material Categorization Performance in Isolated Test Slabs

The first investigation examined the characteristics of various epoxy and grout anchorage systems at the interface between new construction and existing concrete. Twelve different anchoring materials as well as various anchoring methods were studied and compared to a control using no grout. This study does not examine the effects of number of dowels used but was limited to the methods and materials used to anchor those dowels. This experiment was performed on concrete panels in-house at MnROAD. The tube grout method exhibited the best visual and magnetic imaging results. The evaluation methods did not clearly categorize the materials in order of performance but showed advantages of cleaning the drill-hole prior to dowel placement as well as the merits and demerits of using a retaining collar. Results generally suggested the necessity for an actual deployment research project on an actual pavement in real-world service conditions.

Validation in Concrete Rehabilitations

The second investigation sought to validate the findings of the first study out in the field. Twelve different anchoring materials and methods were studied and compared to a control using no anchoring material. The field experimentation and monitoring involved core samples and measured ride quality, International Roughness Index, and Falling Weight Deflectometer load transfer and fault measurement. The control experiment, conducted without any grout or epoxy, initially displayed a notably low Load Transfer Efficiency (LTE). However, over time, there was a gradual improvement, leading to a more consistent LTE, attributed to the deployment of non-mechanical load transfer. Based on the slab thickness, the 1.25-inch dowel did not indicate any statistically significant LTE or other performance improvements over the 1-inch dowel within the anchorage types examined. Overall, the Epoxy Experimental 1 performed best while the un-grouted and unrepaired cells had the lowest performance. Moreover, no anchoring material clearly indicated characteristically low performance.