Assumptions / Promises - Shifting the Paradigm in Systems-Engineering

摘要

In the field of embedded systems, the need for increased competitiveness necessitates the shortening of the development lifecycle. Engineering activities, that in the past were conducted serially, must often proceed in parallel in order to reduce time-to-market. This causes the concurrently working design-teams, to rely on preliminary information that has not yet been finalized. As a result, substantial rework is required throughout the development process, consuming as much as 50% of the total engineering effort. This paper describes early results from the European Commission supported SPEEDS (Speculative and Exploratory Design in Systems Engineering) project. The objective of SPEEDS is to construct theory and tools for concurrent system development which will reduce the overall systems development cost and time to market. We propose a shift in systems development paradigm, relying on the formal use of contracts composed of Assumptions and Promises (A/P), analogous to Bertrand Meyer's "design-by-contract" model proposed in the early 90's. Such assumptions and promises may be related to both technical and management domains, where contracts are defined at all levels of design abstraction. Horizontal A/Ps analysis (Consistency) and vertical A/P analysis (Composability) may be conducted, leading to early discovery and correction of technical or management incompatibilities emanating from invalid assumptions. SPEEDS utilize several enhanced COTS modeling tools, giving users the ability to apply design-by-contract to Model Driven Systems-Engineering in all phases of the development lifecycle.