Saturday, October 22, 2005

Airport Perimeter Security

Leaving Singapore the other day at Changi Airport, I saw different model for security than what seems to be default of a single security checkpoint (e.g. X-ray machine and wand-waving, body-frisking attendants) for all gates.

At Changi, you don't go through security until you reach your gate, each gate has its own security checkpoint. The advantages of centralizing security seem clear - so I started thinking as to what might the advantages of this distributed model.

Theoretically possible would be security customized to the destination, e.g. for flights to political hotspots, full rigour (e.g. laptops get sniffed, every bag gets checked, random body searches, etc) for other destinations, less intrusive security. Travel to such 'safer' destinations wouldn't pay the price of security appropriate to riskier destinations. I don't know if they actually do this.

I saw a specific example of another advantage. As we moved through security at the gate for my flight to Tokyo, the traveller just in front of me was informed by the personnel that he was at the wrong gate. This sort of application-layer check isn't possible with the centralized gatekeeper security model - the gatekeeper doesn't have the application details and so can't apply security based on them.

No comments: