I had the opportunity to interview Professor Alan (Avi) Kirschenbaum from the Technion in Israel about a new project: BEMOSA (Behavior Modeling for Security in Airports). The aim of the project is to improve security in airports through enhancing the capability of airport authority personnel to correctly detect potential security hazards and reduce false alarms. This program is being tested right around the same time of IATA’s “checkpoint of the future.”
SJ: How have the assumptions about IATA’s technology been proven wrong by BEMOSA? Do people really react that strongly when being screened by a machine versus a human?
AK: (1) As the BEMOSA research showed, there are key factors which affect a technology driven security decision and a critical one is an employee’s “trusting” the technology. Our evidence shows that lack of trust (even among screeeners!) leads them to either bend or ignore the rules. At least one-third of the employees do this consistently. These behaviors are due to security officials/policy makers not understanding the social context within which security decisions are made (group rather than individual).
(2) Passengers are and will continue to be screened by a combination of technology and employees as the technology. However, it is vital to understand that technology only provides “suggestions” (looks like a bomb) and it is the security employees who make the decisions. The similar lack of trust in technology by passengers and the human interaction at check points when decisions are made is a unique type of social interaction which we did not study in detail. The ethnographic material we have shows, however, that it appears to be framed as a negotiation situation. The implications of this would suggest that people would rather interact with other people and not be judged only by a machine!
SJ: To create accurate “judgment calls” by security professionals, how much training would be involved for all airports to operate the same?
AK: (1) BEMOSA has shown that the training programs at present are built upon two unsubstantial assumptions that do not match our empirical data garnered from the ethnographic, interview and field survey results. For one, the vast majority of security decisions are made in group context and not by single individuals. Employees consult with one another before a decision is made. This means that training to interpret the technology output (or non-routine situations) specifically for individuals ignores the reality of how security decisions are made.
Secondly, the rules and protocols that are employed in the present training process are limited to a very finite number of scenarios and are reactive in nature. BEMOSA has discovered a practically unlimited number of potential scenarios in its simulation modeling based on the reality of behaviors we observed. This means that future training will need to promote proactive and innovative behaviors rather than rote reactions. In addition, all airports are NOT the same due to physical, demographic and cultural characteristics thereby making it essential that each training program fit the social-cultural context of the potential passengers and employees. Not an easy task but doable.
SJ: Would BEMOSA’s method follow the trend and separate travelers into “normal” “high risk” and “known traveler” or would everyone be treated the same unless a certain behavior triggers the security professionals?
AK: (1) BEMOSA has not dealt with these issues within the framework of its research agenda. However, as I have pointed out elsewhere, selecting/filtering passenger would probably not work as a means to reduce “time in the system” or enhance security.
SJ: Your blog says that technology is a part of airport security systems, but what is the ideal amount or type? The metal detectors we walk through now or enhanced security systems plus BEMOSA?
AK: (1) Difficult to answer definitively! Most of the technology we have (and will have) is based on reactive solutions to what has happened. It’s a “cops and robbers” scenario but from our research, it became clear that the “human factor” will prove to be more decisive than technology in airport security. The reason is fairly simple; technology outputs require human intervention in interpreting the results. It is here that the wide range of human nature, background, past behaviors and characteristics will play a key role in the decisions made. Technology will be part of the background but not at the forefront of the security decisions.
SJ: Where has BEMOSA been tested?
AK: (1) The BEMOSA research project is at the initial testing stage where the simulations and training packages will be put in front of security employees for evaluation and recalibration. The initial tests of the validity of the BEMOSA training program concepts (we call it enhancing security decisions) have been successfully completed and we are now beginning the testing among experienced security screeners (and others). An initial testing at an airport showed great promise but we still need to continue this process.
SJ: What’s the number one reason BEMOSA is better than IATA’s Checkpoint of the Future?
AK: Airports are not mass production facilities but complex social service organizations where employees (and not machines) make key security decisions; the reason why BEMOSA’s program focusing on the security interaction between employees and passengers will provide better security and customer service than one based solely on technology.
Would you prefer security at DIA be based on our social-cultural context or would you prefer a machine to screen you? Or should they be combined?