TRADITIONAL STRUCTURED SYSTEMS ANALYSIS

Traditional Structured Systems Analysis

Traditional Structured Systems Analysis

Hard Systems Vs Soft Systems

Traditionally, real-time systems are classified as being either hard or soft. For hard real-time systems, it is imperative that no deadline be missed, whilst in soft real-time systems it is acceptable to miss some of them occasionally. In practical engineering contexts, the occasional loss of some deadline can be tolerated. This is either because the consequences of the loss can be negligible (i.e. one defectuous part per thousand) or because the robustness of the involved control algorithms imply the ability to react properly at the next invocation step without serious consequences

Not having to meet every deadline allows the size of the system's resources to be smaller than it would be for meeting all of them. This permits the creation of simpler and more cost-effective systems that make better use of the available resources whilst guaranteeing a reasonably good level of service. Hard systems that are not strongly hard will be called weakly hard real-time systems or simply weakly hard systems.

Hard System Methodology

Our minds solve problems by applying a framework of ideas, concepts and methods to an area of interest through our individual perceptions, experience and understanding of reality. Some areas of interest are clearly defined and quantitative problems, allowing us to re-use scientific theories and common themes whilst progressively developing a solution to that problem. Systems engineering is an example of a framework for dealing with these hard problems.

The systems engineering life cycle can been illustrated as a V-model, with the requirement definition and system design activities linked to associated validation and verification activities. The emphasis is on the progressive reduction of risk and the delivery of a balanced system that meets defined user needs. It would be fair to state that the hard systems approach assumes a goal - problems can be solved by the progressive definition and realization of tangible solutions. (See appendix for the Structure Analysis of Hard System).

Soft Systems Thinking

Hard problems are relatively straightforward to visualise, decompose and understand at varying levels of abstraction. However, real world systems include humans and depend on the unpredictable responses and conflicting objectives, perceptions and attitudes inherent with a human activity system. These soft problems are harder to understand and define, so can only be tackled by improving the problem situation rather than by solving the problem outright. Perhaps we could look at soft systems as assuming situations can be improved by human activity and may be supported by tangible solutions.

Soft Systems Methodology

Soft systems methods have developed to take a systemic approach to unstructured problems, recognizing as Chesterton did that it is as important to see the problem, as it is to find a solution. There are several, including ETHICS, Multiview and SSM. The latter was developed to tackle unstructured problems through a holistic approach based on systems theory, recognizing that systems are embedded in a wider human and organizational context.

The original version of SSM is a seven-stage model, although the methodology has evolved ...