Police officers on duty on the road for traffic stops, vehicle collisions, traffic direction, and so forth, are exposed to the risk of being hit or even killed by a passing vehicle. Very few studies have tried to develop a system that can warn pedestrians or police officers on duty on the road to take proactive evasive action. This study proposes an Internet-of-Things protection system for police officers on duty on the road. The development of the system envisaged involves three essential phases: 1. detection, 2. risk analysis, and 3. warning and communication. This study focused on the risk analysis phase. We applied a fuzzy rule-based algorithm that integrates four input indicators (lateral distance from police officer to traveled lane, magnitude of speeding, stopping sight distance, and direct distance) into a single estimate of the risk of a collision. The study used data from a real-world situation on Highway 416 in Ontario, Canada to demonstrate the application of the proposed model. The results clearly demonstrated that the proposed model could generate risk estimates that could be used to give timely warning of a possible collision risk to police officers at work on a road.