Risk scoring: How analytics can improve UK border security

Analytics
Risk scoring provides the best balance of security, cost and passenger convenience

Border management agencies are faced with a constant balancing act between security, convenience and cost. They need to decide whether to carry out checks on huge numbers of travellers, resulting in long delays and inconvenience or – if they scale down the checks – risk letting in travellers and cargo that are illegal or pose a threat.

Attempts to categorise travellers are often ineffective and can be seen as discriminatory if they aren't based on relevant facts. Accurately identifying factors linked to risk is difficult. Powerful tools are needed that can analyse data in real time to highlight security risks.

Tricky decisions

This leaves border agencies and customs offices facing a difficult choice. Do they carry out comprehensive checks, incurring inevitable delays and inconvenience and driving up costs, or do they scale down operations and risk letting in travellers or cargoes which are illegal or pose a threat? In the worst case scenario, more stringent controls can increase delays, without necessarily detecting illegal movements, especially with offenders finding ever more sophisticated ways of evading detection.

Crude attempts at profiling often cause resentment without being effective in identifying high-risk individuals. However, accurately identifying where the risk lies is difficult without tools which can analyse and make sense of all of the available data, and do it quickly.

Risk scoring

So what is the solution to the seemingly intractable problems outlined above? Ultimately, security is the most important issue here and to achieve an appropriate balance between cost, passenger convenience and security, we believe an approach based on risk scoring is the only viable option.

Risk scoring offers an increasingly viable alternative to the traditional security approach of blanket checking all passengers. At its best, it effectively involves using intelligence, behavioural modelling and data analytics to assess the levels of risk individuals pose.

The approach employs complex algorithms and advanced risk management to evaluate whether an individual is a legitimate traveller and as such should freely pass through our borders or be subject to further scrutiny. Rather than leading to more open borders, it can actually significantly enhance protection.

Number of benefits

This risk scoring approach can bring benefits across a range of different applications. It can help identify people who pose a risk based on accurate profiling of all available information, including known activity patterns, watch lists, advanced passenger information (API), containing information like passport number, age, sex, seat number and other data, including where and how the ticket was bought. Analysing all of this together to create a risk score for each passenger not only produces more accurate results, but reduces the potential for inefficiency and resentment which can result from cruder forms of 'traveller scoring'.

Critically, risk scoring can help reduce border queues by enabling border staff to make rapid and accurate decisions about which travellers to question or detain, and which to let pass. Indeed, the approach can enable decisions about whether to stop travellers or freight to be made very quickly. This is especially important when keeping immigration queues to acceptable levels.

Continually tuned

Border agencies also benefit from an approach in which risk models are continually tested against live data and optimised to improve performance and further reduce the number of false positives.

In an ideal scenario, models can be deployed based on advanced and detailed analysis of the information most associated with risk, which are continually evaluated and optimised. By continually optimising the risk model, the company helps ensure that high risk travellers and shippers of high risk freight will be identified even when they try to second-guess the checks that are in place and find ways of evading them.

Agencies can therefore direct their resources to where they know the potentially illegal travellers or cargo are to be found (representing a tick in both the security and cost boxes) and they can also – conveniently – speed up the throughput of cargo and passengers. Combining increased security with a better experience for the great majority of both passengers and freight users is a win/win scenario that makes the case for using appropriate tools highly compelling.

It is time for a change of approach in border management and I believe that risk scoring is clearly the best route forward.

  • Peter Snelling is Principal Technical Account Manager at SAS Public Security. Based in the UK, Peter's responsibilities include raising the awareness of SAS software within the policing and intelligence community.