2 + 10 ≻ 1 + 50 !

Abstract

In traditional design of computer based systems some effort, say 1, is spent on the early modeling phases, and some very high effort, say 50, is spent on the later implementation and testing phases. It is the conjecture of this paper that the total effort can be substantially reduced if an increased effort, say 2, is spent on the early modeling phases. Such a shift in focus of efforts will also greatly improve the overall effects (both quality and cost-wise) of the systems dev eloped, there by leading to a better (denoted by "≻") design process. In this paper, w e specifically consider the design of safety-critical distributed real-time systems. © 2000 Springer-Verlag Berlin Heidelberg.