Andy Davies explores the potentially ‘dark side’ of analytics and explains how it should be used to ensure success and maximise potential.
Analytics has an important role to play in supporting the police to keep our communities safe – and I believe the benefits can be far-reaching. In my previous article (see PP647) I discussed the role analytics can play in policing, reviewed the required data and highlighted a police force that is currently developing its own bespoke solutions.
So where can analytics help?
These are the key areas:
In this article, I will explore the potentially dark side of analytics and outline my views on how analytics should be used to ensure success and maximise potential.
For me, policing analytics can be split into two broad areas:
Resource and demand analytics
Resource and demand analytics should be used with care. Like all disruptive technologies, these techniques can significantly affect the way the workforce looks and moves.
This will raise concerns over the impact on jobs, skills and employment. It is therefore essential to use analytics in an open and transparent way, ensuring that police services can explain and account for their decisions.
It is also important to conduct impact assessments and ethical reviews to make sure analytics is used in a way that is conducive to a positive work environment. Both communities and the workforce need to understand why police use analytics and its positive impact.
Investigative and detection analytics
Police should recognise the risks of using investigative and detection analytics. This type of analytics needs to be used with caution to ensure accurate decisions.
Critics have pointed out bias in some predictive policing models – for example, those that focus on very specific locations – may discriminate against certain groups of people.
A recent report, Data-driven policing and public value by the independent think-tank The Police Foundation, questioned the role of analytics and called for a strict examination of its usage.
The only solution is to use an analytical platform that has a wide range of capabilities and empowers police forces to use their professional judgment.
Police forces have the best understanding of their data, the policing processes involved and the environmental and community factors that are contributing to it. This ensures open and ethical development of analytics for policing and reduces the risk of bias.
The importance of flexibility
The requirements and demands of the community are continually changing, as will the focus of the police. If analytics is to support the police, it also needs to be adaptable. This means moving away from black-box analytical solutions to an open analytical platform.
Platforms such as the SAS suite of solutions – using all the available data – will give the police full control of these processes. They can then use their expertise, experience and knowledge to maximise the impact of analytics to improve the lives of citizens and local communities.
It is important that the platform can give users access to a range of analytical capabilities and techniques in a simple and user-friendly interface across the force, including mobile usage.
Resource and demand analytics used in an open and accountable way will ensure predictive resource and demand management will improve workforce deployment and morale.
Investigative and detection analytics will deliver a reduction in crime with an increased detection capability.
This will ultimately lead to an improvement in the effectiveness of a police force to serve its community.
Vast opportunities in digital revolution
Metropolitan Police Service (MPS) Commissioner Cressida Dick says police must make better use of technology and data if they want to reduce crime and bring criminals to justice.
Delivering the Police Foundation’s annual John Harris Memorial Lecture last month, Ms Dick said as well as creating problems for the police to overcome, the digital revolution also offered vast opportunities for investigating crime.
“A very large proportion of crimes that occur could be prevented, or at least successful investigated in the reasonably near future, by the use of data that is already theoretically available and by technology that is already developed,” she said.
Ms Dick added that detection rates for some offences were “woefully low”, and while the police were very good at making use of technology and data when they investigated the most serious crimes, it was important the same skills and resources were put to use effectively in other areas of their work
Andy Davies, a consultant for police and intelligence services at analytics company SAS UK, says the MPS commissioner was right to say that more widespread use of data in policing was needed.
“It is happening, but it needs to happen quicker and across a wider range of teams,” he explained.
“Analytics is not new to policing. Some police forces have adopted ‘predictive policing’ techniques in recent years. However, to enable the police to respond to the ever-changing demands placed on them, they need to take a more holistic approach to the role of analytics and how it can assist them.
“This means expanding the use of analytics for uses such as demand profiling, resource management and visual investigative analysis. Forces that use data scientists will be able to develop, maintain and put into production a wide range of analytical models that can be used to inform decision-making across policing.”
However, Mr Davies stressed: “Analytics is not a panacea for policing, and it will never be used alone to stop crime before it happens. Officers and staff will remain the most important tool in policing. Yet analytics can help them do their jobs better. We need to find new ways to make analytics and the resulting insights accessible to all levels in the service.
“This is the direction of travel for all industries and policing is no different. Police forces need to equip themselves to handle the large quantities of data available to them, perform quick analysis and take appropriate action.”
Comment