Tamil Nadu Police have launched Spectrum, a risk-based monitoring framework to classify and track individuals accused of sexual offenses.
The system represents a shift toward data-driven policing in India, aiming to strengthen surveillance and enable quicker intervention to prevent further crimes. By categorizing suspects based on risk levels, authorities intend to allocate resources more effectively toward high-risk individuals.
Spectrum utilizes a color-coded system consisting of red, orange, and blue categories to rank the risk levels of offenders [1]. This framework allows police to implement focused surveillance and tailored prevention strategies based on the specific classification of the individual [2].
The initiative is currently being rolled out across 10 southern districts [3]. This region is a priority for the new system because the southern zone registers between 1,500 and 2,000 sexual offense cases annually [4].
According to police data, nearly 15,000 individuals booked in various sexual offense cases have already been mapped across these 10 districts [5]. The rollout follows orders from Chief Minister C. Joseph Vijay for stricter action against such crimes [6].
The monitoring system is designed to provide a structured approach to tracking suspects, ensuring that those posing the greatest threat remain under constant scrutiny. Police officials said the goal is to create a more robust safety net through systematic risk profiling.
“Spectrum utilizes a color-coded system consisting of red, orange, and blue categories to rank the risk levels of offenders”
The implementation of Spectrum marks a transition toward predictive policing and risk-stratification in Tamil Nadu. By quantifying the danger posed by offenders through a visual color code, the state is attempting to move from reactive policing to a proactive surveillance model. The success of the program will likely depend on the accuracy of the risk-assessment criteria and the ability of the 10 pilot districts to maintain real-time tracking of the 15,000 mapped individuals.
