Doyle et al. (1993) describe the decision-making process that led to the choice of a VAV system for retrofit project in an "industrial research facility located in the midwestern United States":
According to the client's direction, each alternative was analyzed based on the Kepner-Tregoe decision-making analysis methodology, which attempts to take into account the wishes and desires of each member of the entire project team, quantify qualitative wants and desires, and reach a decision that best satisfies the whole of the project team rather than one specific department or individual.
The evaluation of key design modifications was undertaken at the study stage in three phases. The first dealt with the desired fume hood sash position at which to re-balance air systems. The system control strategy was the target of the second phase. The third focused on various other independent options.
The decision-making methodology employed forced the decision makers to separate "musts" from "wants." "Must" objectives required that the option (1) improve the safety of the employee (over existing conditions); (2) comply with building codes, governmental regulations, and industry-accepted and corporate engineering standards; and (3) be applicable to other buildings on the campus. Since most of the research laboratory buildings on the campus were originally designed by the same architectural and engineering firms, at about the same time, the HVAC systems are similar throughout the campus, making the third "must" objective a valid desire. If an option fails to satisfy any one "must" objective, the option is not considered. "Want" objectives considered the degree of safety improvement, reliability, ease of researcher use, maintenance operations, environmental (temperature, humidity, etc.) control, energy costs, construction disruption, expandability, flexibility to changing laboratory requirements, construction time, remote monitoring capabilities, use of proven products and technology, capitalization requirements, central heating and cooling plant impact, architectural impact, and reuse of existing systems.
Each team member provided an opinion of the weighting factor for each "want" objective. Skewing of weights was evident for some of them, depending upon individual bias and work function. As a consequence, averaging of weights was used. Next, selected team members scored each option. This was performed without knowledge of the resulting weights to prevent biasing scores based on weights. Again, there were differences between members and averages were used. The total score for each option was the sum of each "want" objective's weight times the option's score for that objective. The "ideal" project would score 1,120 points. It is important to note that financial issues (construction cost, annual operating and energy costs) account for less than 20% of the total possible points! This project represented a dramatically different approach from more traditional system selection methodologies (lowest first cost, shortest payback period, highest return on investment, etc.).
The weighted average diversity of fume hood sash position was 36.5% during weekdays and 17.6% during weeknights and weekends. In addition, we found that some hoods were partially open when no one was actively working on them, so there existed even more potential for higher diversity of fume hood use. Assuming that the unused open hoods were closed to 10% of the maximum open area yielded a potential average fume hood diversity of 16.6% during the weekdays and 11.6% on the weeknights and weekends.
Flexibility and future expansion capabilities [of a constant volume system] are poor. This option scored in the middle ... with the highest energy costs of $770,000 but the lowest capital cost at $1,700,000.
However, a two-position control option was defined and analyzed. This option, which is a compromise between full variable air volume and constant volume, requires a switch or sensor to close exhaust and supply dampers to minimum (non-zero) positions when the sash is less than 25% to 30% open. If the sash is open above this trigger level, full air would be supplied and exhausted.
The projected annual energy savings are $50,000 when compared to the constant-volume option. The two-position option required installation of fume hood exhaust dampers and supply and auxiliary air dampers. Its construction cost of $4,600,000 is $2.9 million more than the constant-volume option and is almost as high as the full variable-volume option.
The main advantage of all variable air volume lies in its capability to match airflow delivery with the needs of each individual laboratory and fume hood. The constant tracking of fume hood use and diversity that is inherent with the variable-volume controls results in a $370,000 reduction in annual energy costs when compared with the constant-volume option. This savings is more than seven times the savings of the two-position option. In addition, the fume hood diversity allows some future expansion capabilities to be easily designed into the system and, as previously discussed, flexibility to changes is inherent to VAVů [T]his option requires the installation of dampers in all supply and exhaust ducts serving the laboratories, fume hood modifications, and variable-frequency drives at the fans. Also, additional controls are required, bringing the projected costs to $5 million, which is $3.3 million over the constant-volume option. The $370,000 energy savings result in a projected nine-year payback of the first cost of the full VAV conversion on energy savings.