Traditionally, consulting engineers wrote their own individual specifications, setting out their requirements for materials and components on each project. Suppliers have long been claiming that this practice has led to a multiplicity of overlapping and often conflicting requirements. Two weeks ago the concrete industry relaunched its standard specification for structural concrete. This week we ask:
Is it time for all engineers to use the same national specifications?
In this competitive age what is needed is a simple performance specification. This requirement was highlighted in 1995 when a study showed that unnecessary detail and restrictions were limiting the efficiency of concrete building construction. The majority of engineers did not rewrite their specifications and industry has therefore produced national specifications.
A specification must convey the engineer's requirement for the present construction, not past problems. The conditions in the specification must be those to be applied, not what might possibly be needed to be relaxed once the contractor is on site.
Relaxation is unfair to tenderers for the specified project, increases costs, leads to conflict between the contract parties and worst of all reduces standards of workmanship.
What is needed is a simple performance specification where the designer states what is required and retains the power to see that he gets it, while the constructor is free to achieve the required performance in any acceptable manner. National documents now being introduced, of which the National Structural Concrete Specification is one, provide this by using a two part document, a base document and a particular project document.
The base document is self contained and held by all parties and does not need to be included with project contract documentation. The engineer completes the project specification, which varies the base document as considered necessary by the designer. This means that the specifier loses no control over the project by using a national specification.
This is to the advantage of all parties; it is a win-win situation: - The engineer does not have to write and keep up to date full specifications, but is free to concentrate on particular project requirements. The constructor gets a familiar specification and can assess the risks knowing that he can use familiar techniques as well as innovation. The client gets a well built project on time. Engineers must adopt national specifications and standards.
Taking specification writing out of the engineer's hands would be disastrous. Several reasons spring straight to mind. Sitting down to write a specification one is obliged to run through a mental model of the construction process. Unexpected problems are not exactly unknown in the construction industry and this is a point at which an engineer speculates on what might happen and provides for it.
Making the engineer write the specification probably has more benefit for the engineer than for the specification. This is to the greater good of the industry. You have to know your stuff to write them. A standard spec ought not but can be used by anyone. And if the specification is an original then the contractor has to read it. We live in the real world of commercial pressure. How many specs get their first real airing at claims time?
Overspecification is the likely outcome of using a standard spec. If such a document is universally adopted then everything will tend to be too good for its purpose. The whole basis of design is economy, which is the same thing as only reaching for the quality which will just do.
Notes on a drawing or a document looking like War and Peace; which will get the better price?
Underspecification would also happen. Thomas Eads engineered the first major steel bridge in the United States. He learned his stuff building iron-clads in the Civil War so he knew his material. He was stubborn as a mule when it came to the steel he would use and tested every bit delivered to site. The supplier, Andrew Carnegie, said he would rather have made the bridge from silver, given the quality demanded.
Eads did the industry an immeasurable service. His quality level became the benchmark and provided a firm basis for the phenomenal developments which followed. It should be engineers who establish the standards not the supply chain.
The best work is always by the pioneers, not their followers.
Pioneers are by definition operating beyond industry standards.
What does engineering idealism have to do with prescriptions?
Engineers are not legally obliged to refer to British Standards and codes of practice in their specifications, or to check that any standards or codes of practice they do cite are still current.
All readymixed concrete is batched by weight, yet many structural concrete specifications still include mixes such as 1:2:4 or 1:3:6 which are based on volumetric batching.
Most structural codes of practice specify no more than six or seven grades of concrete. Readymix depots routinely supply 20 or 30 grades, with characteristic strengths differing by 0.5MPa or less.
The concrete industry estimates that widespread adoption of the new standard specification could save £30M a year.
Among those already backing the new specification are clients BAA, Stanhope and consultant Ove Arup The National Structural Concrete Specification is divided into three parts. First, it identifies clauses for standard concrete construction. Secondly, it provides clear examples of the specific requirements for individual projects. This highlights 'by exception' the required amendments to the standard requirements in the first part. Thirdly, comprehensive guidance to the specification's philosophy is provided.