Information Technology Reference
In-Depth Information
Failure Modes
•…….
•….
•..
Noise Factors
People
Environmental
Operating System
Customer Usage
Coupling
Response FRs
User
Software
Input (M)
FR
Control Factors
Ideal Function
β
M
FIGURE 18.7
P-diagram.
may start only after 10 seconds and, finally, may not start at all. We want to minimize
the variability in the output response to noise factors while maximizing the response
to signal factors.
Noise factors are those factors that are not under the control of the software design
team. In this television example, those factors include speaker conditions, weather
conditions, battery voltage level, and television wear. Signal factors are those factors
that are set or controlled by the customer (end user) of the software to make use of
its intended functions.
The goal of a DFSS optimize phase is to find the best experimental settings of
factors under the team's control involved in the design to minimize quality loss;
thus, the factors in the experiment represent control factors. Signal, noise and control
factors (design parameters) usually are summarized in a P-diagram similar to the one
in Figure 18.7.
18.5.1
Usage Profile: The Major Source of Noise
A software profile is the set of operations that software can execute along with the
probability with which they will occur (Halstead, 1977). Software design teams know
that there are two types of features: tried-and-true features that usually work and fancy
features that cause trouble. The later is a big source of frustration. End users will
Search WWH ::
Custom Search