Search for an answer no longer!

Although I am not a security specialist, I do have a numerical background and can provide the following analysis of the problem which might help John and any further readers wanting to find an answer to this dilemma...

The best way to understand the statistical nature of searching is by way of example. Consider a venue that will accept 100 visitors, with five of them carrying something unacceptable such as drugs. The first realisation is that the only way to stand any chance of preventing all carriers entering the venue is to search everyone. Even if 99 out of 100 people are stopped and searched, the 100th may well be a carrier.

The whole issue is one of probabilities. The question to ask is: ‘What is the probability of finding a carrier of prohibited materials if one-in-five is searched, one-in-four is searched, one-in-three is searched and so on...?’

Table 1 (below) shows how the probabilities of catching 0, 1, 2, 3, 4 or 5 carriers during the search process change with search frequency. As the number of searches increases, the probability of catching all five unwanted entrants rises (but even when searching three out of every four individuals entering the premises, the probability of all five carriers being stopped is only 23.7%).

Due to the low probabilities of all five carriers being caught, the conclusion must be that searching alone provides a very low probability of completely preventing the entry of prohibited materials.

For several search frequencies, Table1 also shows the probabilities of only one and, one, two and only 2, three and only 3, etc carriers being caught. If all of these probabilities are added we begin to see the useful figure of the probability of one or more carriers being caught (as Table 2, below, duly indicates).

Thus even searching as infrequently as one in every five, with five carriers in 100 visitors there is a better than 2/3 chance of catching one or more of them. However, if all that happens to a carrier when caught is that they are refused entry to the venue then they are likely to chance the odds. After all, from Table 1 there is only a 0.03% chance of one or more carriers gaining entry.

As an individual, with a one-in-five search, there’s an 80% chance that person will not be caught. For the penalty at hand, these are odds that may be accepted.

On the other hand, if someone is caught carrying something that’s unacceptable in law and they are then detained, handed over to the police and generally have a thoroughly spoilt evening, they may think twice about trying! After all, as an individual there is a 20% chance of being detected.

The conclusion? Even at a modest frequency, searching has a good chance of catching some carriers if there are any.

However, if the response is to make an example of those caught, it is the example that is likely to be more of a deterrent than the physical catching. It is always worth remembering that objective analysis – and, importantly, a detailed consideration of the post-detection actions – should facilitate an optimal search, detection and response process on site.

Dr Bob Anderson MBCS BIFM (Qual), Independent Facilities Management Consultant