This is surprising. Electronic data interchange, or EDI, is constantly hailed as a great advance. Documents can be sent almost instantly, amended immediately, stored and retrieved with ease. And there is a wide range of products on offer, from inexpensive Internet-based posting applications, such as CADWEB, to Bovis鈥� mighty Hummingbird system, which tracks and controls documents throughout a project.
It is this range that is the problem. All of the products on the market do something slightly different, work slightly differently and, worse, use different languages. Documents written in Lotus Notes are not easily readable in a Microsoft Office application such as Excel 鈥� or even in an earlier version of Notes. The same situation arises with drawing software: for example, AutoCAD files are difficult to transfer into MicroStation.
This is why Hawkins says the programs are creating more problems than they solve. And he should know 鈥� G&T is working on 30 projects that use some form of electronic data interchange, from simple e-mail to full-on document management. Eleven of these projects use specialist project management applications, and they are all different.
What do you mean by that?
On one recent scheme, the specialist software for transferring and managing CAD files did not take into consideration the different scale conventions used by the architects and the engineers. This was disastrous. The drawings produced by the two professions have to be brought together, which means that they must be layered and numbered to one standard that can be used on both AutoCAD (for architects) and MicroStation (for engineers).
If everybody works to the same original, the lead designer can overlay all that information from the CAD drawings. That can be a big if. On the G&T job, the two sets of drawings did not tally, with the result that Hawkins had to print them out and go through the measurements one by one.
And his experience is not unique. Mervyn Richards, senior computer-aided drawing and design consultant at Laing, suggests the problem is endemic to the design team: 鈥淯nfortunately, some architectural practices and engineers are not interested [in working with the same conventions],鈥� he says.
Another drawback of having multiple systems is the cost, both up front and hidden. Simple applications such as CADWEB are not expensive to buy, but when several applications are needed, the cost mounts up. Some of the more sophisticated systems also involve training costs. Two years ago, Ove Arup Partnership was quoted 拢2m to install an independent system that would not even be able to read all the programs it uses. instead, it decided to develop its own system for transferring drawings, which it will be giving away (1 October).
Problems also arise when staff are transferred from project to project and have to learn new programs. Gleeds Technology managing director David McAll says: 鈥淵ou get on to a system and you have to understand the common interface. It鈥檚 like the difference between WordPerfect and Word,鈥� he says.
The hidden costs include the time it takes users to get up to speed on several systems. Time is also wasted if drawings have to be re-measured because the layering protocols are not standardised, or if document files have to be re-sent because they are in the wrong format.
One firm has become so exasperated with the plethora of applications in use that it has decided not to use any of them. Rory Bergin, IT manager at HTA Architects, investigated the possibilities of using a specialist document management system but decided it was too expensive. He is now e-mailing documents in Adobe鈥檚 Acrobat. This is written in Postcript, which can be printed but not edited. Obviously, this means that its uses are limited, but all file formats can be read, so there are no worries about interoperability. The system is also cheap, at about 拢140 a licence.
For their part, the software houses that offer communication systems insist that although there are many incompatible ones, they are better than nothing. Ray Crotty, managing director of C3 Systems, acts as a consultant to firms setting up document management solutions. He says that on a typical 拢50m project, every 拢1m of construction turnover generates 400-500 drawings. He believes that technology can control them much more effectively than relying on the post, paper and filing cabinets. He says problems have arisen only because the technology is still relatively new: 鈥淧eople are just going to have to grin and bear it.鈥�
Do projects need information managers?
Grinning and bearing it is, however, not enough for G&T鈥檚 Hawkins. A team from G&T, Davis Langdon & Everest and the RICS Construction IT panel are looking for a grant to investigate the way electronic data management systems are used. Part of the research will involve assessing the need for a full-time information manager on projects to control documents and ensure that everyone involved complies with set standards. The team has just been refused a Partners in Innovation grant from the DETR, but Hawkins is determined to carry on. 鈥淚f I have one crusade left in me, it is this,鈥� he says.
Laing has also attempted to overcome the teething problems. Richards has compiled a standard layering system for CAD drawings on all projects and a standard numbering system for drawings and documents. The contractor has also developed a system for transferring drawings, Dislist, and uses data management system PC Docs. Richards is now starting to suggest to all clients and project teams that they establish a set of standards before projects get under way.
Standards are the key. Lack of standards makes document management almost impossible. But these cannot be formulated until the industry decides what it wants and how to get it. The IT departments at consultants, architects and contractors around the country are battling to bring some form of order to electronic project management in individual offices but, like Laing, they are all coming up with different solutions.
As time goes by, market forces will weed out the less popular products, and there will be some degree of convergence, but it is anyone鈥檚 guess how much. Possibly, certain products will emerge as market leaders for a particular size or scale of scheme. 鈥淎s in the car market, there will be competition, but BMWs don鈥檛 compete with Minis,鈥� says Gleeds鈥� McAll. But for the time being, Hawkins believes the industry is trying to do its business in the manner of an earlier, catastrophic construction project: 鈥淚t鈥檚 like the Tower of Babel,鈥� he says.
