Linking the Construction Industry: Electronic Operation and Maintenance Manuals: Workshop Summary (2000)

Mr. Ken Poirer

Tweddle Litho Company

Today I hope to draw some parallels between the automotive and the construction industries. Some of these will be obvious; others will stretch your powers of imagination.

A key to this effort is keeping the building professionals and the data professionals focused on their respective fields, yet work together. We think the best way for data professionals to help is for the building owners to identify their goals and needs. Data professionals can then develop ways to achieve those goals.

The Tweddle Litho Company does a wide range of publishing-related activities within the automotive industry. We write the copy in Standard Generalized Markup Language (SGML), print the book, and put the same information out on compact disk (CD) or on the company's Internet. We provide foreign language translations and print from as few as five manuals to as many as one million.

Tweddle Litho services the automotive, medical, chemical, and electronics industries. We publish operator and owner guides, service and repair information, training manuals, directories, parts books, price lists, and other publications.

We handle all owner literature for the Ford Motor Company. For Daimler Chrysler, we produce all owner, service, diagnostic, and flat-rate manuals worldwide. At General Motors, we produce owner manuals and other miscellaneous publications. For Nissan, we handle owner, service, and flat-rate manuals. In some cases, as with Nissan, we offer the information as well as produce the literature.

When producing domestic manuals, in some instances, the authoring source (customer) exports SGML to the Tweddle Litho database. In other cases, we have local authoring sources that will supply that information to us via the Internet, T-l lines, or on a disk.

Tweddle Litho converts the data using conversion software. This involves taking the SGML and putting it into an ASCII format that allows other software to compose pages. Once the pages are composed, we send electronic files back to the customer for approval. When we receive approval or edits from the customer, we make the edits, recompose the pages, and send them through a proofing cycle. Once the proofing is done, we send that information to be printed or uploaded to a CD or the customer company's Internet.

Tweddle Litho also produces foreign language manuals, which involves basically the same process. Once we receive the SGML files, we can compare them with a previous file and identify changes. The changes are sent to in-country foreign language translators, the translated changes are integrated into the previous file, and then returned.

The automobile industry developed a set of standards for the electronic interchange of service and diagnostic information in response to the Clean Air Act of 1990 (J2000 standards). I think the T2008 standard for trucks is more applicable to the construction industry. It involves heavy trucks, which typically have more maintenance and are on the road for as long as 20 years. Maintenance is tracked and updates are made when different components are used for replacement or repairs. The parallel with the Operation and Maintenance Support Information (OMSI) system is to create a standard for all the operation maintenance and repair manuals. This ensures uniformity of delivery and consistent data output. Every book should have the same look and feel, so people can share that information from one building to the next. They also want to share that data from one book to another book to optimize the data.

What are some of the similarities between the two industries? The use of standards, uniformity of input format (SGML), the deliverables, the ease of use, trackability, and long-term data usage.

Where do these industries differ? At present, the construction industry is not as suited for a complete change in how information is provided. Not everybody currently has the resources to provide information in SGML. For the automobile industry, it is easier because it has the resources. The only challenge for automotive companies is to find a common format.

For buildings, the intervals at which information is updated might not be as frequent as with vehicles. Also, although the automobile or trucking industries supply technical service bulletins to their representatives in the field to show changes, there may not be a comparable application in the building industry. If building equipment manufacturers are not organized to provide updated information on building components, updating the information provides a challenge.

The publisher's challenge is to take every conceivable format in which the construction industry is going to provide that information (Pagemaker, SGML, Word, WordPerfect, Portable Document Format (PDF), hard copy, handwritten, etc.) and narrow it down into Document Type Definition (DTD)-compliant SGML for the creation of the manuals and CD output.

How do we bridge the gap? How do we create a solution? You have to think outside the box. The easiest way to create the solution, but which may not be cost effective, would be to give everybody an SGML monitor to provide that information just the right way from the data professional's perspective. We need to make it simple for everyone. We need to allow for a variety of formats, but at the same time reduce the possibilities. We can create data templates for non-SGML data using the lowest common data formats.

Another possibility is to give direct Internet access to allow anyone to enter data online, using a template. Once the information is typed into a template, it is invisibly tagged, and SGML makes it applicable to that DTD. Some of the recommended solutions are to develop a free or inexpensive forms-based authoring tool. That might be the template for the different Word or Pagemaker files. Another solution is to outline a limited set of alternative data formats, provide authoring templates, and develop conversion services.

Updates to the data can be made in SGML by the vendor. The vendor can create a database of manufactured products and can pull from that database any duplicate information. If one submits a suggestion, if they are using a common format from one building to the other, there is no need to recreate that data. Once data are converted, they can be used over and over again.

Here is a work flow vision: Receive the data in multiple formats, convert them to SGML as necessary, validate the SGML files against the DTD to make sure that the file is usable once it is

converted, run the SGML files through conversion scripts and through a composition engine to compose pages in final format, send the proofs out, and upon approval, print manuals and output CDs.

One of the next steps is for the printing and the building industries to determine the total number of formats in which the information can be provided. What is the lowest subset of information that needs to be provided that can be turned into SGML?

Another step is to determine the viability of creating a free or inexpensive authoring tool. The creation of templates and conversion scripts for SGML data conversion is needed. To determine how this will work, the publishers could receive sample or test files in order to create sample operation manuals and do a pilot test.

I'll leave you with the following questions:

• How can SGML data be used beyond the printed manual and the CD?

• Can it be interchangeable with a CMMS system?

• Can the data be downloaded from the operation manuals directly into a CMMS system?

• How will updates be handled?

• What systems will be in place to maintain the integrity of the data and what are some of the future applications of those data?

PARTICIPANT: Where are the automotive industry and your company going with respect to XML?

MR. TAKACH, Tweddle Litho Company: We have stayed with SGML because it is the standard the automotive industry has chosen. XML is a subset of SGML, and we can convert the SGML to XML later if needed.

PARTICIPANT: One concern that I have is that many craftspersons and technicians are so accustomed to paper documents, stick files, and drawings that there will have to be cultural change. We must create the technology that will enable all of this to happen, but that will not occur if we do not at the same time work the cultural change that is required at a personnel level. A second concern is that we need to take the data in our CMMS and generate valuable information for the facility managers to make informed decisions on how to use their resources.

MR. KATZ: Twenty years ago the collective bargaining units of the various craft unions would refuse to work with or would demand special compensation to work with keyboards and screens. That has changed. It started shifting about 10 years ago when, in the negotiating sessions, the unions demanded to be the ones to use the electronic tools, because they knew if they didn't their members eventually would have no jobs. As far as management goes, managers also have to be electronically enlightened, or they are not going to have their jobs anymore. That is where the culture is today.

PARTICIPANT: I agree; however, is it not equally important to develop capabilities to convert these data to information that a facility manager can use?

MR. KATZ: The conversion of the data is what this is all about, and as this grows, there will be a defined cost and a defined flow, just as we are living with scanning paper documents into CAD. What does it cost per square foot to scan paper documents into CAD? A dollar or a dollar and a half a foot? Sometimes the cost of scanning data into CAD exceeds the cost of the building of 50 years ago. We are going to be doing two things. We are going to be funding and converting existing data when the dollars drive it, tagging the existing data, and marking the existing databases. More important, we will put a requirement on the software systems that are creating the databases. The CMMS will tag the work order data and the inventory transactions. By running the CMMS system, you are automatically tagging the

operation information. Then the standard management tools, data reporting tools, will work with all brands.

MR. TAKACH: In the automobile industry we had the same problem 10 or 15 years ago. We addressed it by continuing to provide paper manuals. We slowly moved the mechanic from a paper product to an electronic product by just not making the paper available anymore. Now, for example, Nissan is going to go to a complete electronic manual by model year 2001.

The mechanics have reacted to this by recognizing they can print off a page from the electronic manual if so desired and take it over to the car and work with it. Most of them have adapted and said, “I will use my computer because the information is right here on the screen, and I can just keep screening down or tabbing down until I get to what I want.” It's a matter of education and working with people long enough to make sure that they feel comfortable working with the electronic media.

PARTICIPANT: My observation is, “Do not underestimate the importance of training.” So as you are bringing in the new technology, be generous with the training of the staff that you want to use it.

PARTICIPANT: Can the problem of getting the information into CMMS databases be solved with SGML? What are the manufacturers doing? It seems to me that they should be leapfrogging. Instead of just going back to their old manuals and tagging everything, they should be thinking about how to develop tools to put the data into an electronic form now, as they develop the products. For example, pick a couple of systems and make sure that the available data automatically goes into the computer systems to monitor or maintain facilities. Are you covering that issue in the work done to date? Do you have the manufacturers involved? Are there some manufacturers way out there leading the pack?

MR. BRODT: We have some of the manufacturers here today that we have been working with. Principally, we have worked with Johnson Controls, Siemens, and Trane.

MR. FORD: Johnson Controls is represented here because we believe this is the direction in which the industry is going. Right now, there are a lot of standards coming together in an ad hoc manner. Various manufacturers are trying different formats. We would like to see the industry follow common standards.

PARTICIPANT: I wonder if you could comment on the building automation portion of the industry and the whole idea of interoperability. Do you see the standard extended that far? Do you see a single, seamless database at that point?

MR. FORD: Yes, we see all of that coming together into a common system.

MR. KATZ: When used on data, SGML can in a very smart, powerful way identify a specific, as fine a granule of data as is needed. We are marking, we are tagging, we are flagging data, and it goes across all brands of software, and it even works on mainframes. That is the power to think about. It is the data that are tagged, and we as managers, buyers of the software, have to put the requirement on the software producers to put in the tags. The people who own the data force the suppliers to open it up, so you can go actually deal with another vendor.

MR. RICE: I came here today because I actually am designing and building the kinds of systems that you are talking about. The question is, “What real use is an SGML document? If I give you an SGML document, what can you do with it?” The answer to that, by itself, is not very useful at all. SGML represents kinetic energy. That means I can turn it into a form that can be processed by a CMMS system or some sort of automated maintenance browser system.

The key is that the information is only as smart as I make it. It is only as smart as it is tagged. Just creating a simple SGML document will not allow me to take that information and plug it into a CMMS system and have it do something useful. If it were that simple, then we could all just take HTML pages off the Internet.

SGML is more difficult than some other ways to create information. The payback is that you can do more things with your information, and you can put it into more forms that have more uses.

Whatever you discuss in this workshop and whatever you incorporate into the DTD that you eventually produce, the utility that you get out of your information will be directly related to the level of intelligence built in. Unfortunately, the ease with which the information is produced is proportional to how intelligent it needs be.