Success factors for cost effective computerization
Editor’s Note: System integration offers major economies in managing the information flow in and between organizations. The integration of administrative systems has been in process for at least two decades. The integration of technical systems has been developing more slowly and primarily in volume production systems. The integration of administrative and technical systems poses an even greater challenge, especially on project type work.
As a result of concerns raised through the Construction Industry Institute, research has been undertaken to understand what integration of technical systems in the construction industry really means. This article is a summary of the most important findings on this issue. For more details, contact the authors at FAX: (415) 642-1246.
The nature of the issues raised in this article provide insights as to the comparable issues in project areas of other industries.
Many people in Construction believe that computer technology is and will continue to be an important means for improving Industry cost effectiveness. A number of design, construction and management functions, typically referred to as Computer Aided Design (CAD), Computer Aided Engineering (CAE), and Computer Aided Construction Management (CACM), have already been computerized, Much still remains to be automated. The rapid rates of change of technology and the expense of employee training jeopardize those investments though.
Although computers are widely accepted, there is significant concern that adaptation is not progressing in a cost effective manner. In response to this concern, we investigated the cost effectiveness of computerization in large owner and contracting companies. This article is one of the products of that investigation. Other reports in this series document computerization trends and a framework for investment analysis [1,2].
Findings presented here were collected mainly from large owners and constructors that are members of the Construction Industry Institute (CII). Most of these firms are of dominating and influential size and, hence, should be watched carefully for indications about the entire contracting industry. We hope these findings will assist computer users and managers throughout the Industry to use computers more adeptly and avoid some of the costly mistakes of the past.
Many factors affect successful computerization. Listed in order of importance, the following five were identified in our study as most important:
- Management Commitment
- System Planning
- System Standards
- System Integration
- Training and Education
The following sections elaborate on these factors.
Computerization to date in Construction has been driven mostly by sophisticated users and enthusiastic middle management. Many users have proven the cost effectiveness of computerization to themselves and forward- thinking top management; yet in some companies others are still struggling to receive support from senior management.
The majority of the benefits available from this bottom-up driven computerization have already been realized. That is, users have automated the methods by which they do their own immediate tasks, The next advancement in computerization will be through integration.
To achieve integration, organizational barriers must be bridged. This requires higher-level management direction. Executive managers must recognize the importance of their philosophy and leadership in achieving successful computer integration. Commitment and support are required to create the atmosphere that fosters creativity and innovation. More management direction and involvement is required to set goals and timetables. As one Industry expert aptly expressed it, “computerization is management.”
Wide disparities exist in different people’s skills in using computers. Consequently, management must start with an analysis of employee computer literacy. People problems normally rate as the top priority. Technical problems are usually of second level importance today.
To address this illiteracy question and other issues, many companies have established core groups that manage overall computerization efforts in the firm. People in this group must have a solid understanding of the engineering- construction business and work processes in addition to new technologies. Common responsibilities include:
- Identify user needs and establish priorities for system acquisition and development.
- Select, maintain and publish both inhouse and industry standards. These standards should include short descriptions of programs and systems, developer information, and required hardware and software to run the system.
- Manage new system development by enforcing an approval procedure.
- Coordinate system development and provide quality assurance.
- Check conformance to company and industry standards.
- Promote use of new, cost effective vendor and in-house products.
- Publish newsletters to share information among company users including industry trends, success stories, solutions to common problems, reviews of new hardware and software products, etc.
Like any other project, good planning is crucial for system development. Long-range, yearly, corporate-wide and project level planning are mandatory for successful computerization. The following issues must be addressed periodically at these levels: hardware, software, people, productivity, schedule, budgets, and intraand inter-company standards in the Industry. Database, networking and communication are important issues to be considered in system planning.
Generally, it is best to involve all interested parties, especially potential users, in the planning stage. Though the development process tends to be slowed with so many people, it is a more thoughtful and team-oriented effort. When system requirements are identified early by planning participants, extensive modifications are later minimized. This is becoming more crucial as computerization tends toward more integration. Work scope, level of documentation, and on-going support need to be determined clearly at this stage.
Unlike the previous decade when most software had to be developed inhouse, good software is now abundant on the market. Third-party software can deliver benefits by saving in-house development cost. Long term, it is important for the Industry to promote healthy competition among software vendors by supporting them through buying and using their products.
System compatibility will be improved if the Industry roses more third-party software. Software vendors who offer the best system compatibility will be Favored and eventually survive, The Industry then should share experience through various forms of user groups to guide third-party vendors. One Industry expert compared the case of each engineering company developing its own software to a hypothetical case of every construction company making its own construction equipment. This analogy underscores the importance for engineering-construetion companies to keep their focus on building services and not software development.
Bottom-up driven computerization has resulted in many isolated automaion islands with no standardization to speak. These islands can be based on organizational, hardware, data, user or a variety of other reasons. Although this diversity has historically been required to meet the needs of different business environments, it adds substantial system support, training and upgrade costs. Without a well-structured set of standards, system integration, if it is even possible, is costly and the exchange and reuse of information is difficult. System standards are prerequisite to system integration.
Establishing standards and implementing them effectively will be a challenge especially for large organizations because it is more than a technical problem. People will resist changing from familiar systems, They like the idea of establishing standards as long as they are their own. Compromises mean having to give something up. Selecting the right standards and implementing them properly requires good management direction. If the management mandates a standard too early or forcefully, it may end up with a standard for each system rather than one universal standard for all systems.
A key group of people must be responsible for preparing, maintaining and promoting the use of company standards. Trends in both the Construction and Computer Industries need to be closely monitored to ensure that company standards are developed consistent with others. The companylevel team may be best positioned to participate in broader forums.
Formal procedures are required for developing and documenting any system. Underlying assumptions, qualifications) limitations, and input and output media need to be clearly outlined in this documentation. Desired features that are not yet avai1able but soon will be can also be listed for further enhancement so that others can improve systems at a later date. This listing will also provide indications of future system directions.
Integration is probably the single most important key today for achieving cost effective computerization. Many companies indicated that the next major advancement will occur through integration of various project functions and different disciplines both within and between engineering and construction. Integration means different things to different people at different phases of a project. For our purposes we developed the following as a working definition:
Integration provides a common database that is accessed, used and updated by multiple applications or users. The information in an integrated system is organized in a logical way and demonstrates a centralized behavior with consistent and-redundant data.
Integration allows for multiple, separate databases that can interact with each other transparently to the end user. One end-product of this technology “weaving” is integration of engineering, procurement and construction work processes. People will have to be more broadly skilled.
Integration delivers many benefits. Being difficult to quantify these benefits makes true investment analysis quite hard. In part this is because so many of the advantages occur over along period of time and within a complex, interdisciplinary and dynamic company. Among the important ones are:
- Improved data integrity and quality.
- Timeliness of data.
- Elimination of redundant data collection and entry.
- Improved communication.
- Distributed processing results in less dependence on physical location and allows more flexible work assignments.
- More opportunity for inter-company interaction among the owner, engineer, vendors and contractors on a project.
- Easy reuse of data on current and future projects.
These benefits translate, at least indirectly, into reduced cost and improved schedule, which means improvement in productivity.
Priorities and strategies to achieve successful integration should be to:
- Establish clear goals and objectives.
- Understand necessary information flow; possibly different from existing document flow. Islands with mostly bi-directional information flow are first candidates for integration.
- Define integration scope by defining boundaries of logical application islands.
- Identify opportunities for combining application islands by bridging and maybe merging.
- Secure cooperation from all involved parties. Both top-down and bottomup commitments are required for successful integration.
- Best results come from thoughtful analysis and design. Patching and linking rarely deliver the best longterm results.
Integration between different departments creates severe organizational problems. It involves bridging traditional departmental boundaries. Ownership, type of access and format of stored information need to be clearly identified because different departments use different formats, types and abstractions of data. Responsibilities and priorities in maintaining, updating and using centralized information need to be specified precisely. Integration should be directed by upper management because of these cross-departmental ramifications.
Training And Education
Effective computerization has to start with improving the time lag between what people use and what would be best to use. Training and development of people is especially important in keeping computer users current with rapidly changing technology. Motivating end users to learn and continuously improve their computer skills is a major challenge to management.
Training at the wrong time can be wasteful. Computer knowledge becomes obsolete in a relatively short period of time. Training must be continuous and a part of company culture. Community college and video-taped courses are being used by some companies for this training function. Reward mechanisms for motivating people to self-educate is also used in many organizations.
Project and corporate functional managers need training to understand their roles and the importance of computerization. Production personnel must keep their technical skills up-todate. In large companies which can afford the expense, in-house training is generally more valuable than training by vendors. Adequate resources must be provided for training programs. Some suggestions for employee training are:
- Provide regular training programs for basic skills and general purpose systems and as-needed training programs for specialized application systems. As-needed training programs should be established based on the input from system users (trainees) as well as from the group who provides training.
- Use a Corporate Computer Information Center to plan and coordinate system purchase, use, development, maintenance and training. This Center would:
- Provide guidelines to ensure that upwardly compatible software and hardware are purchased or developed.
- Manage new system development to avoid duplication of effort.
- Support standards.
- Implement procedures to maintain system compatibility and consistency.
- Use incentives for employee development, such as:
- Employee recognition through issuance of certificates.
- Portable PCs and modems for home use.
- Paid-hours for training. Some firms match these hours to employee contributions to discourage paidhour abuse. Firms in other industries let employees spend 10% of a work week for self-determined research, study and personal improvement.
- Use established education and training programs by vendors and local colleges.
OTHER SUGGESTIONS FOR COST CONSCIOUS SYSTEM USE
A number of other important issues and suggestions for better system use were collected in our study. One common idea was public recognition of the importance of people in developing and using computer systems. All the companies in our study sample stress the critical Importance of efficient computerization. Actual computer hardware and software systems are secondary to the people. Good people can always make the best use of the system they have, but good systems alone do not produce good results.
Further suggestions are summarized along four major lines: People, Software, Other System and Vendor-related.
1) People Issues
- Ensure that development, enhancement and usage are driven by the engineers, designers and others who actually use the systems, as opposed to computer specialists.
- Training, documentation and support for users can rarely be over-emphasized.
- Have engineers and designers use the CAD systems as creative design tools, in addition to drafters using them as “Electronic Drawing Boards.”
- Build users’ confidence in computer performance. Once confidence is lost, it is very difficult to restore. Effective training and maintenance of system performance are important.
- Management leadership by top-down example setting is crucial. Day-to-day use of computers by top managers will give them a sensitivity to the power and capabilities of systems.
2) Software Issues
Interface to Materials Management
- Work to develop “intelligent” drawings that contain non-graphic information or are attached to databases to yield bills of materials (BOM) and other management reports as a byproduct. View systems as a tool for information management, in a true project management spirit, not just as “dumb” drawing generators.
- Integrate materials management programs to avoid duplicated data entry, ensure consistency and eliminate associated errors.
- Provide data communication links between materials management programs and other applications to automate and streamline work.
- Provide back-up systems of material management databases for critical construction activities. Loss of materials data for critical activities will be disastrous.
Interface Between Preliminary Engineering and Detail Engineering
- Use basic 2-D/3-D graphics instead of plastic models to optimize conceptual and detailed planning work. Resolve problems with these basic computer models before employing more elaborate plant design software.
- Maximize use of preliminary engineering work by preparing planning studies in a format upwardly compatible with rigorous plant design software.
- First, standardize the upstream design work, such as flow sheets, P&ID’s and equipment lists, to Facilitate integration between upstream drawings and downstream detail drawings. Then concentrate on later design activities and then functions of construction.
Cost Effective Design and Drafting
- Develop libraries of standard details and standard drawings. Different sets of standard drawings can be stored for repeated use in different project phases.
- Use similar drawings from previous jobs as “go-bys” to minimize original work.
- Produce 2-D drawings and other construction documents from 3-D computer models. If 3-D models cannot be accessed and viewed directly from the jobsite, use video-tapes prepared from a 3-D model or dynamic walk-through simulations to communicate with field personnel and to incorporate their input for constructability improvement.
- Scan existing drawings into CAD compatible formats instead of digitizing or redrafting because of the cost.
- Use “Autodraw” facilities to tie design calculations to technical programs and databases automatically producing many types of simple drawings.
- Allow engineers to do “night work” at home using company-supplied com puters and modem connections.
Use of Centralized Databases
- Work toward one-time-only data entry through effective information management.
- Use and apply standard specifications and codes to the database.
- Include construction and maintenance information in the database.
- Plan database security and back-ups carefully.
3) Other System Related Issues
- Maintain responsive systems with adequate peripheral capacity, functional file and disk management, and performance utilities.
- Use a variety of micro and minicomputers on a network environment. Rely on file servers to avoid redundant storage media. Ensure proper back-ups and archives.
- Maintain electronic communication of drawing files and data between internal offices and with client offices.
- Increase use of engineering CAD models and walk-through simulations and what-if’s in construction planning and control in the field.
- Provide guidelines, procedures and utilities to facilitate drawing productions.
- Provide user commands, menus, cell libraries, tutorials and other utilities for automatic generation of graphic elements to avoid creating them every time. These utilities facilitate managing drawing files and databases. They also expedite interference detection, drawing extraction and annotation, and material takeoffs.
4) Vendor Related Issues
- Maintain good relationships with vendors to ensure proper support and attention to practical aspects.
- Support employee participation in user group forums.
- Promote data exchange standards, perhaps through joint ventures of users and vendors.
PROBLEMS AND NEEDS
Two major impediments to faster and more efficient adaptation of advanced computer technologies are: 1) lack of Industry-wide standards, and 2) system incompatibility.
This study revealed that today’s single most important Industry need is establishment of standards. Industry’s standards development efforts are weak and not well coordinated despite recognition of their importance. Their absence results in companies spending an enormous amount of time and money duplicating vendor catalog data for their own use; e.g. piping components, instruments, structural shapes, etc. Moreover, these products are constantly updated requiring ongoing renewal of existing databases. In this context, note that standards refers to both the technology used and the engineering- construction procedures common to the business. Attention must be given to both aspects of this issue.
Relatedly, the absence of data transfer standards aggravates the incompatibility for CAD drafting standards, media standards and product codes, and vice versa. Engineering documents created by different systems cannot be used or transferred causing expensive and unnecessary re-creation of those documents. The Construction Industry must impress the importance of standards on software vendors to maximize electronic file transfer, to minimize the manual creation or re-creation of data. Defining a standards methodology will be more important than a symbology. For instance, a piping and instrumentation diagram represents a methodology; a valve icon represents symbology.
Industry representatives should take the lead in this regard and suggest a set of standards to vendors that maximizes electronic file transfer. Leadership by owners in the standardization effort will be crucial for success. Forums such as the PMI publications are one means by which these needs could be voiced. Use and support of a neutral data interface format by the Industry is important and should be pursued now. Establishing universal Industry standards will take time due to fragmentation. Nevertheless, the banking and grocery industries provide examples that it can be accomplished.
The need for establishing standards falls into four general application areas:
- Data transfer standards (i.e., direct translator or neutral files),
- CAD drafting standards,
- Media standards (e.g. communication, storage, etc.).
- Product codes.
Industry-wide support and acceptance are required for all four areas. With an exception for the first category, which is discussed in detail in the following section, the Industry’s coordinated and directed effort on standards is presently minimal due to the following reasons: 1 ) conflicting vendor- vested interests; 2) varying level of details required for different project types at different design stages; 3) contractors themselves with conflicting interests; and 4) antitrust and other legal concerns.
Construction must find the means to implement the cost effective use of a neutral data interface format, such as Initial Graphics Exchange Specifications (IGES) and Product Code Exchange Standards (PDES), Because it will take time for the computer industry to come up with universally compatible systems, the use of a neutral data interface format is necessary to achieve cost effective use of computer systems. Strong support from Construction for establishing standards, including active input to the IGES and PDES groups, is required. That is not happening presently. Standardization will probably require that owners take the initiative since they have so much to gain.
Other problems and needs in cost effective computerization were identified in our study and are listed below:
- Means to improve compatibility between systems and with existing drawings.
- Means to improve poor vendor software quality and after-service.
- Better materials management and control, including maximizing justin-time delivery to minimize material storage and handling costs. This will be achieved by increased computerization among all involved parties.
- Software and hardwdre selection criteria and performance standards.
- Effective system implementation and training procedures.
- Cooperation between Industry and universities in education, research, and training.
- Means to positively influence the computer industry, including software system vendors.
- Cost effective organizational structures for computer support.
- Standard accounting practices for cost recovery, both internally (between departments within an organization) and externally (between different organizations such as the contractor and owner).
- Acceptable means for progress measurement in computerized design and engineering. Number counts of completed drawings are often used as a progress measurement parameter. This parameter will be much less valid when computerized systems are more fully used and fewer drawings transmitted.
- Powerful and inexpensive computer hardware and 3-D software.
One inherent problem stemming from the Industry’s fragmentation is the difficulty of motivating engineering contractors to spend additional sums for state-of-the-art computer systems, particularly when design and construction are performed by separate entities. The design organization in these cases has no share in “downstream” field cost savings, including those accruing from savings in the owner’s operation and maintenance costs. Indeed, additional engineering work to maximize downstream benefits can often result in more engineering costs, which can make the engineer less competitive. The owner’s role and the appropriate contracting pattern in this situation can be seen as crucial to achieve overall cost effectiveness.
The new business pattern of owner-contractor partnering may be a viable solution to this problem. Turnkey contracting should be seriously considered whenever additional investment for computerization in engineering has a potential for significant downstream benefits. In other words, a desire to capture the maximum possible benefits from computerization is another persuasive reason to partner with full service constructors.
Computerization is more a management than a technical problem. Until now most automation has been bottom-up driven. This trend is reversing to a topdown effort. Executive management must now recognize the importance of its role in computerization. Strong management support and commitment are required for overcoming organizational problems. Effective computerization will demand system integration going beyond traditional organzational boundaries. This will require substantial investments and patience.
The next major advancement in computerization will take place by integrating computer systems through databases that behave in a centralized manner. Integrated systems will permit multiple data handling requiring data input only once. Electronic mail communication through networking will advance in the near future as design and construction processes are integrated. Lack of Industry-wide standards and system incompatibility are identified as two serious problems for productive computerization. User-oriented groups must take the lead in voicing the Industry’s specific needs and preferences to the system vendors.
One powerfully clear message from our study is that users accomplished successful computerization when they focused more on the people issues than the technical details. Competent and timely training of management personnel as well as system users is crucial for successful automation. A computerization strategy should be tailored to reflect the specific needs of the firm. The most sophisticated hardware and software systems alone will not deliver cost effective computerizattion. It is the people who will deliver that value.
The writers gratefully acknowledge the support of CII and their member companies for sponsoring this study. Paul Cooper, Ben Hoff and Paul Teicholz deserve individual recognition.
1. Choi, K.C. and Ibbs, C. W., January 1990. CAD/CAE in Construction: Trends, Problems and Needs. Journal of Computing in Civil Engineering.
2. Choi, KC. and Ibbs, C. W., June 1990/Costs and Benefits of Computerization in Design and Construction. ASCE Journal of Engineering Management.
C. William Ibbs is an Associate Professor of Civil Engineering at the University of California, Berkeley. He has been actively researching computerized project management issues for a decade and used them in corporate organizations as a project engineer before that. Current interests involve expert and object-oriented database research for cost and schedule control.
K. C. Choi is a doctoral candidate in Berkeley’s Construction Engineering and Management Program. He is a research engineer for the Automated Integrated Methodologies Group at Bechtel, Inc. His previous affiliations include Bechtel Petroleum and Dillingham Construction.