In the early 1990s the UK government noted an emerging use of computer-based technology in applications with safety implications. The Government funded a research programme, SafeIT, to look at the implications of this technology in various industrial sectors and particular with the early work going in to the standard that became IEC 61508: Functional Safety.
One project co-funded by SafeIT and the automotive industry was MISRA which developed a set of guidelines for safety-related software development in the automotive context. The MISRA Guidelines as they became known, which were first published in 1994, were notable in codifying for the first time the emerging principles of IEC 61508 for an automotive context. Some 15 years later in 2009, ISO released the draft of ISO 26262 the automotive variant of IEC 61508.
Although the publicly-funded part of the MISRA project ended in 1995, members of the MISRA consortium agreed to continue working together. One of the first results of this ongoing collaboration was the release of the first version of MISRA-C in 1998.
Since then, partly due to the global take up of MISRA-C, a second version of MISRA-C was published in 2004 and additional documents have been issued covering automotive software readiness for production, safety analysis, the model-based development or Auto-Code generation series and MISRA-C++ with other work and documents in the pipeline for 2010 and beyond.
Part of the UK Government SafeIT project co-funded with the UK Automotive industry in the early 1990's MISRA began work on a set of Guidelines for the use of software in automotive environments. These were the some of the earliest work in codifying standards for vehicles along the lines being deveoped for safety critical use in the IEC 61508 functional Safety Standard This was longbefore the automotive variean ISO 26262 was even started. The Guides published in 1994/5 were intended to have a 10 year life span ie until about 2005 bridging the gap until IEC 61508 came out. In fact IEC 61508 came out between 1998 and 2000 in seven parts.
The companies making up the MISRA group were:-
The Guidelines were published in 1994 with 9 supporting reports published in 1995. Hard copies are no longer in print but PDF's can be obtained from the MISRA website. The Guidleines were also became and ISO Technical Report (TR15497) and can be obtained from ISO The supporting reports contain more detail and rationales to explain the recomendations in the Guidline document
The document provided assistance to the automotive industry in the application and creation within vehicles of safe, reliable software.
This report covers the aspects of vehicle engineering that relate to software to support integrated communications and diagnostics networks. The subjects addressed in the report are for the whole vehicle architecture, communications and multiplexing, on-board diagnostics, off-board diagnostics, tools and testing. The intention was to review the wider aspects of automotive engineering which can directly influence the software that is embedded in vehicles.
The recommendations and processes givne in the report were designed to enable a developer to determine a system's integrity level and therefore adopt a development process in order to achieve the confidence level required in the software. The recommendations could also be used by assessors of these systems to help determine if the methods used by the developers were suitable for the intended application.
The report covered issues arround electromagnetic compatibility (EMC) and the implementation of real-time systems. The report assumed the hardware hasdbeen designed to reject electromagnetic interference and only considerd what additional requirements could have been taken in software.
.The report explored the use of software in the design of vehicel control systems. It was divided into three parts: theoretical considerations,
design considerations,
practical considerations.
The report identified a number of software attributes and metrics that could be used to provide usefull data and therefore gice an indication of the quality of software.
A modern vehicle is made up of many subsystems, many of which will use software. The report presented the verification and validation activities that should be performed upon subsystems and particularly the software subsystems, and of course on the complete system. The verification and validation activities were given in the logical order they would normally be applied to for a typical vehical lifecycle. Specific methodologies and techniques were detailed in appendices.
Not authoritive on the (ever changing) legalities of writing contracts, hoever, the report was intended to be an engineer's view of the topics which should be considered by engineering teams and purchasing people involved with software products buying, selling, creating, managing. It sets out to explain why the topics covered should be considered. Bearing in mind that software is not somethign that can easily be weaighed and measured as most items in an automotive environmet can be. The report was logically divided into three parts: software aspects of subcontracting in the automotive industry, technical aspects of contracts and commercial aspects of contracts.
The Human Factors Engineering implications of MISRA are considerable. Its influence as a discipline can be seen in many of the stages of the life cycle.
It contained of a list of reference and source documents used during the project. It also summarised the findings of the initial phase of the MISRA project.