Monitoring workplace safety: Nonlinear optimization in a statistical model

Abstract

Given a cluster of enclosed work areas, the problem is to maintain frequent monitoring of the atmosphere in each area for hazardous levels of contaminants. A central mass spectrometer is capable of doing this monitoring, provided it can be used efficiently. Mathematically, a mass spectrometer may be modeled by a general linear statistical model where the unknowns are the concentrations of the possible contaminants and the observations are the gate readings of the number of ions with a chosen mass/charge ratio. The accuracy of the observations depends on time, and the necessity for frequent monitoring imposes a time constraint. The mathematical problem can be formulated as a problem in the theory of optimal experimental designs. This paper will present a theoretical solution to the problem, an algorithm to implement that solution and some empirical results obtained from the implementation.