AXIOMATIC DESIGN IN SOFTWARE DESIGN FOR SIX SIGMA (DFSS) - Software Design for Six-Sigma: A Roadmap for Excellence - page 336

TABLE 13.2 Software Design Parameters (DPs) Examples

Design Parameter
Considerations DP Example

User Product design for user profile.

Subject-matter
expert

Product design for consistency with expert opinion.

Designer Reflection of product.

Customer Reflection of customer preferences beyond product.

Functionality Individual independent tasks performed by the product.

Integrated functions Combined tasks necessary to complete a transaction or other

function.

Menu User interface display permitting access to features.

Domain Coverage of specific information.

Equivalence classes Determination of inputs generating a consistent product

behavior.

Boundaries Parameters where product behavior is altered.

Logic Sequence of actions following a consistent pattern.

State-based Use conditions indicating different function availability or

product behavior.

Configuration Ability for product to work in different intended operating

environments.

Input constraints Determine how user or system can enter data.

Output constraints Determine how data or information is displayed.

Computation
constraints

Determine how data is computed.

Storage or data
constraints

Determine limitations to data.

Regression Impact of incremental design changes on the product.

Scenario Complex fulfillment of a particular set of tasks.

Business cycle Scenario intended to replicate product use for an entire

business cycle.

Installation Initial application of product in its intended operating

environment.

Load Ability to handle excessive activity.

Long sequence Sustained product use over an extended period.

Performance Speed and duration of product use.

Comparison with
results

Determination of variations to external references.

Consistency Determination of variances to internal product references.

Oracle Comparison to common acceptance indicators.

extreme situation, A could be a complete, that is, nonsparse full lower or upper,
triangular matrix. For example, in a full lower triangle matrix, the maximum number
of nonzero entries, p ( p − 1) / 2 where A ij = X = 0for j = 1, i and i = 1 ,..., p.

A lower (upper) triangular decoupled design matrix is characterized by A ij = 0for

< > >

i j (for i j ). A rectangular design matrix with ( m p ) is classified as a coupled

design, as in (13.3).