The diagnosability properties of a sensor network are defined in terms of fault observability, fault resolvability etc. However, it is difficult to compare different sensor networks based on these metrics as the indices are inherently incommensurate. Hence, algorithms for sensor network design are essentially minimum cost designs subject to complete observability or maximum resolvability. Also, the sensed data in a typical chemical process application can be used for material accounting and process control, apart from fault diagnosis. Thus, sensor network design involves multiple objectives, designers and decision makers. Hence, development of an integrated sensor network design algorithm that addresses all these objectives simultaneously is an important activity. One common approach to solving such multi-objective optimization problems is to develop a utility function. We have quantified the utility function or value of a sensor network from a fault diagnosis perspective in terms of the expected value of the operating profit. A two step procedure for determining the value of a given sensor network from a fault diagnosis perspective has been proposed. In the first step, the set of resolvable faults for the given sensor network is determined. Subsequently, the value of the sensor network is determined by calculating the expected value of the operating profit. Evaluating this quantity exactly is cumbersome, both analytically and numerically. Hence, reasonable assumptions inspired by process knowledge are made to determine an approximation to the overall value.