Page 52 - Industrial Ventilation Design Guidebook

3.2 DESIGN METHODOLOGY DESCRIPTION I 7
The first and most important aim of design methodology is to produce, by
systematic analysis, a description of the design procedure that is commonly
accepted and used in every process in different markets. The idea is to make a
description of the technical process of design; in other words, to answer two
questions:
* What is to be made clear and done during the design procedure?
» In which order are the tasks to be done?
The design methodology does not take a position on who does this or that
task. That is part of administrative or commercial flow, which varies in differ-
ent parts of the world and even in different projects in one country.
Once described, the design methodology can be applied for several usages.
The need of the design methodology description was identified at an early stage
of the Design Guide Book project. As such, it serves as a clue for this book. As
design methodology vigorously describes the connections and correlation of
different tasks, it could also be utilized as a skeleton of computerized design
tools, such as expert systems, and as a tool for systematic analyses. It can be,
and already has been, applied in research projects in order to identify gaps in
knowledge or by financiers when assessing the need for proposed research.

3.2 DESIGN METHODOLOGY DESCRIPTION

Basic elements in the methodology can be presented in several ways. Table 3.1
gives an idea of the whole contents. In addition, decision trees are needed, be-
cause the design process requires many back couplings which cannot be illus-
trated in table form. The decision tree technique is a tool for dividing a
process, here Design Methodology, into subtasks, which have their accurate
inputs and outputs. The order of the tasks is chosen so that the data needed to
do a task are given or calculated before that task to minimize the number of
back couplings. Thus, the tree guides the right execution order of the subtasks.
It also serves as an internal quality guidance tool for design process, because
the quality of the preceding subtasks' results will be assessed in the next task,
where they are used as input data.
In real projects the number of back couplings is much higher because of
the administrative process, where the accuracy of the input data will improve
during design. In early stages of the design one has to work with very prelimi-
nary and inaccurate information to produce preliminary results, such as cost
estimates and space reservations, which have to be given in spite of missing
data. Nevertheless even in such cases the whole decision tree must be gone
through; the only difference is that missing data are replaced with an estimate.
When missing data become available, the process is gone through again in or-
der to review the plans. Such descriptions that cover all possible back cou-
plings cannot be made, because completion of the projects varies among the
different countries, branches, and partners that are involved. As a matter of
fact, the nature of targeted design emphasizes the core idea of the design meth-
odology. To achieve the optimal solution and all the original targets, one has
to redesign all the tasks following the point at which a change, such as new in-
put information, was made.