In very large infrastructure projects, there is no clear delineation of discrete steps as there is in this illustration. Design and engineering continue while construction is underway, responding to conditions in the field. While the infrastructure is in operation, additional projects involving design and modification or new construction are initiated. To put it simply, everything changes at the same time. An information technology solution must take this constant and simultaneous change into account; traditional GIS software is basically a standalone function that cannot take the changes in other areas of the project and cannot be incorporated on an ongoing basis in the other information technology involved in the project.
There are many IT and Web solutions today, but none of them addresses the big picture. As a practical matter, resolving the bigger picture is a lost cause. CIOs find it a challenge just keeping the partial solutions in synch while adding productivity to the entire E/C/O (engineering, construction and operations) process. In most cases, there is no "in between project time" when entire systems can be implemented and workflows reorganized.
Nevertheless, the pull of the Web is strong, and during the dot-com gold rush all sorts of would-be "killer apps" appeared on the Internet, and many lessons were learned. The first wave consisted of customer-oriented portals (B2C). The required transaction models are quite simple to program: For online shopping, it would be selecting a book online, entering secure payment processes and closing the transaction.
The process required to manage the construction of infrastructure facilities across extended geographical areas is on the other end of this business model spectrum. It is, to my mind, the biggest challenge to move to the Internet. Road construction is a perfect example.
The complex factors and constraints for the management of large infrastructure projects include the following:
- Very high-cost projects
- Liability risks at every step
- Very long duration (years) for each project
- Multiple participants (transportation agency as the owner-operator, various federal agencies, local government, private contractors and consultants, and the public)
- Extensive and strict regulations to comply with (from environmental issues to bid letting)
- Massive amounts of data, recorded and stored in many disparate forms (from index cards to digital satellite imagery)
- Accuracy required in the smallest details of this information
- Overlay of front-office (geoengineering, project management), and back-office (enterprise and ERP) processes
- Politics (both the "good" democratic process as well as the inevitable internal politics within large public and private organization)
Clear project objectives
Despite the complexity of large infrastructure projects, the project objectives can be stated quite simply: Reduce cost, shorten time, lower liabilities (This was the message of Dr. Kam Movassaghi, Secretary of Transportation, Louisiana Department of Transportation and Development during the Area II HEEP (Highway Engineering Exchange Program) meeting in Baton Rouge, Louisiana on May 30, 2000).
Another goal, which promotes all three objectives, is to catch the design errors early. The US Department of Defense estimates that the cost of correcting a mistake in the design phase increases by three orders of magnitude-a factor of 1000-if the error is detected late in the construction phase as opposed to early in the design phase.
Many project managers understand that underlying all four goals is this requirement, perhaps the most difficult of all: providing the right information to the right person at the right time. This is more than just giving the person tools to search for the information, as he or she may not be aware of its existence. What is needed is a comprehensive, sophisticated engineering information management system.
Where is the killer app to manage this complex process and reach these simple goals? The "mother of all process management system" does not exist, nor will we see it during our lifetimes. If the Web offered a single, miraculous solution, we would know it by now.
Some pieces of the solution are growing on the Internet currently. Developers recognize that one of the major hurdles for engineering information management is the complexity of current business networks. Today, they are maintained as well as can be done today with mail, telephone calls, faxes, E-mail messages, travel and overnight shipment of documents. As the mother of all complex networks, the Internet seems the appropriate upgrade for all these disconnected means of communication and information management.
Network of networks
During the life of an infrastructure project, a variety of communications and business interfaces arise. Using a biological metaphor, these "membranes" form the boundaries between the participating bodies (people, organizations, companies, constituencies). The nature of these membranes also changes in time (a bidding company may become a contractor; individuals will change positions or leave). In terms of the Internet, these membranes represent the B2B, the B2G (business to government) and the G2G interfaces.
Many processes across these membranes can be Web-enabled, and the first ones to have been deployed are communication and information-posting. In each such case, a "portal" or E-mail hub has been identified and is owned by the leader of that process (with or without being hosted). The result for large projects becomes a complex Web of hubs and spokes.
Although this gives the impression of a powerful communication network, in fact, it is nothing more than a large number of small, independent systems. The server-to-server layer is still missing. Server-to-server is one of the hot markets today (for example, Microsoft just launched its new NGWS architecture).
Among the many requirements of IT for Infrastructure construction and geoengineering data, perhaps the most important is the necessity of managing individual design elements at a low level of detail or "granularity" (down to the man-hole or the curb and gutter). These are pay items and this is where the errors can occur … or where they can be avoided.
There are other issues somewhat unique to the Internet that must be addressed in any Web-based solution. One is security. Although paper-based information may be susceptible to copying, theft and destruction, these issues are addressed on a local basis. There is little concern that someone on the other side of the globe, who has access to a computer and too much time on his hands, is going to expend much effort to copy, steal or destroy paper drawings, film and file cards. The Internet creates a form of intimacy not only between trusted associates, but with untrustworthy and even criminal elements all around the world.
A second issue is bandwidth. While it may be inefficient to ship documents and images and look through them manually for information, slow Internet connections in some regions and among some participants in infrastructure projects makes the use of the Web even less efficient. There are emerging technologies, however, that allow rapid searches of massive data without transmitting all the underlying data across the Internet. Also, there are now techniques for panning and zooming across large image files, where the server does the heavy lifting and sends only what the client, the Web browser, needs at each moment. For an example of this technology, test drive Bentley's ModelServer Imager at www.bentley.com.
Finally, even beyond the issues of bandwidth associated with conventional telecommunications, there are those in the field who need access to the information and query tools. Again, wireless communications is seen as a high-growth area, and many new approaches to getting data out of the office and on construction sites are being introduced.
Even with the promises of these new server-to-server communications, project managers are well aware of the human factor in these large processes. There will be a limit in the return on investment when very large support systems are deployed.
The best approach is to continue identifying discrete business processes and improving their performance with a mix of business-process re-engineering and Web-enabled technologies. Keeping an eye on the future, just be sure that the technology you adopt is adequate to allow for the eventual linking of all the servers together.
Conclusions
The Web today is used in two ways. First, the hub-and-spoke model, where content providers post information for a defined set of users (which could even be the public at large) to access remotely, essentially in a read-only mode. Second, peer-to-peer networking essentially carried out by E-mail (this is the Internet "phone system"). Although these two models offer benefits, they are not adequate for the management of large projects, where information represents the most valuable asset.
In the world of infrastructure projects (from small ones to billion-dollar projects), the paper processes used today have yet to be reproduced on the Web. But just over the horizon, we see the beginnings of global engineering information management: data creation, project tracking, peer-to-peer communication, actual project teaming, all being managed at the server level, and assuring that the right information is delivered at the right time to the right person.
The many-to-many interaction - across the network of networks - is the challenge. Project collaboration extranet portals are a beginning.
But in the E/C/O world, in real projects, with contractors, subcontractors, surveyors and disconnected users, the information is being created simultaneously in many places. The back-office management system is the key to attain new productivity levels. Many ASP technologies are available now. The best ones are able to adapt to the user workflow and also maintain a private area (on LANs) next to a more public area where contractors meet. This is one of the clear objectives of Viecon, the Bentley online project collaboration technology available today.