Solution House uses AI to predict crime

Read time 2min 40sec
Solution House Software director Tiaan Janse van Rensburg
Solution House Software director Tiaan Janse van Rensburg

Cape-based urban management, public safety and security solutions provider, Solution House Software, has added a module to its urban management system, which uses artificial intelligence (AI) to predict and map potential crimes before they happen.

The urban management system, called Incident Desk, was launched last year. It provides facility, security and urban managers with real-time incident alerts via a mobile app.

The new Predictive Analysis module uses machine learning technology developed by Solution House, together with aggregated data from multiple information sources, to determine the likelihood of different types of criminal activity in the Incident Desk management area.

Incident Desk is used by companies that monitor numerous buildings, or areas such as estates, neighbourhood watch groups, shopping malls and schools.

"With the module installed, Incident Desk generates seven- and 30-day forecasts as heat maps based on crime types and incident probabilities that managers can use to optimise their finite security resources," says Solution House Software director Tiaan Janse van Rensburg.

"Crime is notoriously difficult to predict, but given that Incident Desk can access so many different types of data - including weather patterns and forecasts and historical data - the results are based on fairly accurate and proven trending algorithms."

He says the algorithms used are based on international academic research and specifically developed for crime.

The module will initially focus solely on crime, but future versions may be expanded to include other industries such as facility management and maintenance.

Janse van Rensburg says the technology is self-learning, so it will continuously review, correct and refine predictions. This means predictions may become more accurate over time.

He says the module has been used to correctly predict crime.

"The system provides a probability (high, medium or low) that a specific type of crime will happen in a specific area on a specific day and time of day. The following day or week, we or the customer check the predictions against the actual incidents of crime and we have had numerous hits."

The predictions will only be made available to managers and their support staff.

"The type of information generated by the module is highly sensitive, and is therefore limited to restricted personnel," Janse van Rensburg says.

"Incident Desk will function normally - with residents in a managed area still able to report incidents by phoning their security service provider control room or through the smartphone app - but information like crime analysis and prediction will be strictly delivered to the relevant management personnel only."

The company plans to integrate the module into all current users' systems as a free trial. It will be phased into a subscription model next year.

The Predictive Analysis module will be rolled out to all current South African multi-tenant customers towards the end of this month.

See also