Port Technology discusses the latest advances in container security. Securing the supply chain and scanning cargo containers is a complex issue with significant policy impact. One of the most prominent items of regulation in this respect is the US SAFE Port Act of 2006, which introduced the requirement to scan 100 percent of cargo containers entering the US. But, even in the year 2012, six years after its introduction, the original intention of ‘Security and Accountability for Every Port’ was still not compatible with technical and organisational realities. Consequently, in 2012 the US Department of Homeland Security deferred the requirement of 100 percent scanning until July 2014, using a foreseen exception option in the SAFE Port Act. Therefore, the challenge of improving container security remains high on the agenda.

The European Commission’s Joint Research Centre (JRC) has worked on the issue of container security for more than 10 years now, starting with mechanical and electronic seals. But recently a paradigm shift has been introduced to the research activities: instead of trying to protect the traditional container itself, new container concepts and new monitoring approaches were looked at, and first research results are promising. The requested range of risk management tools was pursued in four major research strands, which in the future might support a comprehensive, multitechnology, and agreed-upon approach to container security: (i) new container materials; (ii) smart electronic sensors; (iii) wireless multi-hop communications, and (iv) mathematical outlier detection and security analysis. While the first three of these research areas are connected to the physical container itself, the outlier detection is a mechanism applied to ‘metadata’ (itinerary, transhipments, etc.) that does not have to come close to a container to perform a security analysis. Read the full article – Future cargo containers: how smart can they get? here!.

In contrast to the physical container technologies (described in the article), this outlier detection approach does not target the containers themselves but addresses the metadata information available about the container (e.g.itinerary or transhipment data). JRC has built a container traffic analysis system (CONTRAFFIC) which mathematically analyses the itinerary data of cargo containers, identifies unusual stopovers or loops and enables the research analysts to detect suspicious containers from a distance. The mathematical algorithms of CONTRAFFIC are continuously being improved, connecting more and more data with each other and analysing more than five million containers per month and matching the data mathematically against 800,000 import declarations. The system is currently queried over 20,000 times by different EU customs per year.

A first test of the system in a practical exercise with customs authorities has produced hit-rates of over 30 percent on a very much narrowed-down set of suspicious containers which finally were inspected. It needs to be underlined that the JRC’s CONTRAFFIC cannot be seen as a tool to identify all risk containers, and it does not even claim to identify a significant share of them, but it provides an efficiency gain for the inspectors and enables them to add additional information to their cases or to identify new cases before they take operationalaction. Insofar CONTRAFFIC is not meant to find the needle in the haystack, but to identify some of the many needles in a haystack with a higher probability. Source: Port Technology