WSP07134

• File Number: 8283.200
• Description: Colorado River Decision Support System - CRDSS
• State: CO
• Basin: Colorado Mainstem
• Water Division: 5
• Date: 1/1/1993
• Author: Dames and Moore
• Title: Feasibility Study Report for a Colorado River Decision Support System
• Water Supply Pro - Doc Type: Report/Study

Table 11: Estimated Costs of Database Design and Development

Person $8,500 I $12,000 I $13,000 I
Task Months person person person
month month month
Slation DB 6 $51,000 $72,000 $78,000
Physical Characteristics DB" 42 $357,000 $504 .000 $546,000
Historical I Near Real-Tune 26 $221,000 $312.000 $338,000
Model DB 10 $85,000 $120,000 $130,000
Policy DB 21 $178,500 $252.000 $273,000
User DB 5 $42,500 $60 ,000 $65,000
Tolal 194 $1,649,000 $2,328,000 $2,522,000

Estimating the costs for integrating the necessary models into CRDSS was the
most complicated 'pan of this pan of the study. The reason for this is that many
uncertainties exist regarding the nature of the existing model implementations
at the source code level. To arrive at a reliable cost estimate, a mixed
alternative cost approach was applied, First, the costs for a "best" model
alternative were estimated (Table 12). This alternative was based on a selection
of what were considered the "best" model alternatives for serving the various
broad need categories, Next, other alternatives were fonnulated by replacing
one or more of these "best" models by other available models,

For each alternative the following cost categories were considered:

. User Interface Costs, Depending on the existence and quality of
the user-intetfaces supplied with the models, additional intetface
work will be needed to embed the model in CRDSS,

DMI Costs, In a data-centered approach individual applications are
connected to the central database (and possibly each other) through
a Data Management Intetface (refer to Task 5 for a more detailed
discussion ofDMIs).

Model Enhancement Costs, Depending on how well the
capabilities of existing models fit the needs as expressed in the
needs analysis (Task I). models may have to be extended and
enhanced,

For all of these three categories costs are likely to be higher for the more
complicated models than for less complicated models. It was concluded that
the cost estimates for the various alternatives mentioned above exhibit very
minor variations, Costs for the inclusion of nine models corresponding to nine
broad need categories will take an average of 166 person months (Note: tables
for the replacement model alternatives are not shown here, They all resulted in
very similar cost estimates).

.

.

e

Modeling Costs

e

e

DAMES& M OORE/CADSWES.47