Abstract

A public works programme requires that a number of more and less hazardous tasks are being performed by a number of different people over a period of time. Now we wish to minimise the risk to the workers involved in the programme. But what does this mean? Even if we constrain ourselves to the risk of death, there are multiple interpretations of the ideal of risk minimisation. Do we wish to minimize the number of expected deaths amongst workers? Do we wish to reduce the risk that is imposed on the most vulnerable workers? Do we wish to minimise the chance that even one worker dies? Do we wish to minimise the chance that more than an acceptable threshold of deaths will occur? These all constitute different ideals of risk minimisation. When we invest in risk minimisation, we need to take into account certain technological constraints. Some investments may be highly effective in that they make substantial reductions to the risk involved in certain tasks, whereas other investment decisions may be less effective in doing so. We construct models to determine how we should allocate investments to reduce the spread of risk, given a particular technology and given particular ideals of risk minimisation.