Performance specifications translate user requirements into a technical form to which industry can respond. (2 slides.)

In our work to develop good lighting in schools, we have found a need for low brightness and low contrast in the lighting, so that the blackboards can be seen without glare and the printed pages can be read comfortably.

In this case we set forth a series of performance requirements, and when they were presented to the lighting industry, half the industry said the standards could not be met because they contradicted the laws of physics-which was a pretty strong condemnation of the performance specifications. The other half of the industry quietly designed systems that are not doing the job effectively.

We developed a combined lighting and ceiling system that covers large areas of the ceiling. The system requires that there be a capacity for acoustic absorption in the ceiling and that air be able to move in and out of spaces.

By developing performance specifications, a method is evolved for translating user requirements into a technical form that industry can respond to and this type of vehicle for translating needs into technical terminology is needed.

If industry is to gear up and produce, industry needs specifications which are understandable, which can be tested, and through which industry can develop and test products and can show that the producers will in fact meet the carefully stated needs.

In many cases, before sending performance specifications out to bid, it becomes necessary to be sure that there are ways for industry to meet the requirements. Therefore we have done our own design work internally, to develop components that will prove the performance needs can be met, before we go out to bid.

The tooling and the manufacturing capability of industry invariably provide better ways of doing jobs than we have conceived-once we go out to bid. But first we have to show how the job can be done.

In other cases we indicate through diagrams a range of ways in which a given set of performance specifications can be met, then we find that manufacturers will be able to bid who represent a wide range of materials-one using timber, another steel, and another concrete all to meet the same set of specifications. In saying what the job should be rather than how it should be done, we have found that we can get a tremendous range of competition.

Now, as we start a new building program we know that we must be concerned with the entire process by which buildings are builtfrom the use of raw materials at the start, right down to completion of the buildings.

For instance, in a recent housing program it was our purpose to design a kit of parts that can be produced effectively, to provide a large variety of different housing types.

The program for doing the job had to be well ordered, and so, as in our other work, a critical path flow diagram had to be designed showing how all of the parts would go together. Then it had to show how just the smaller subsections would fit together at the building siteto do a total job within an assembly process.

In that case we are working with pieces and parts, as they are purchased, as the components are put together into modules, and as those modules are developed into buildings. Because of shipping, in some cases you may ship the components to the site. In other cases you may want to have modules at the site. This will vary with each different situation.

As we start a new building program we must be concerned with the entire process by which buildings are built, from raw material to completed building.

But how the sequence is arranged provides a control of all the pieces and parts so that they will all work with one another and do a total job, and so that the job provides for appropriate variations to meet needs, and so that entire process will move quickly, meeting a given set of user requirements.

We were working in a student housing system, now being tested by the University of California, where well over 30 different manufacturers designed and bid unique products.

The system developed includes a new type of concrete structure. The wet members of the structure are perforated, and the services needed in the buildings can be moved in two directions through this rather unique type of concrete structure.

In developing this material and system, we have come up with some particularly interesting technical developments. For example, the concrete members are pierced to provide penetration through them for mounting air conditioning, piping, conduit, and other service components. In testing the strength of the material, the Portland Cement Association found that the concrete members were made stronger when regular holes were cut through the members. In this extremely interesting discovery, we learned that the process of cutting the holes increased the strength by 25 percent. We then found that we had been preceded in this discovery by the aerospace industry, where a group had been testing frames to achieve a particular structural strength. Finding that the frames failed in the same place each time, the technical staff finally yielded to the derisive cries of the custodian of the laboratory, who said "cut holes where it fails."

After many failures, one of the engineers cut the holes, and the structure did not fail. Everyone crowded around the laboratory custodian and asked why. His comment was, "Well, you know, toilet paper never tears on the holes." An interesting thing is that with the concrete members it took 7 years for this piece of information to be transferred from the aerospace industry to the building industry— accidentally.

In the structure I am describing we were going for long spans to provide for flexibility in the arrangement of space. We had also developed an air-conditioning system that was designed to be compatible with the structure. One of the interesting things here is that a building system was designed through the process that saved about 8.5 percent in the first cost of construction. Much more will be saved in owning costs of the buildings over time-very substantially more. It is hard to tell at this point just how much. In addition, the performance has been improved because the average span will be raised to 30 feet from a nominal 14 feet, so that the spaces provided in the building would be extremely flexible. In these buildings the partitions are 100 percent demountable, and each student has complete control of his or her environment.