The cost of large systems' Verification, Validation, and Testing (VVT) is in the neighborhood of 40% of the total life cycle cost. The cost associated with systems' failures is even more dramatic, often exceeding 10% of industrial organizations turnover. There is a great potential benefit in streamlining and optimizing the VVT process. The first step in accomplishing this aim is to define a VVT strategy and then to quantify the cost and risk associated with carrying it out.
This paper provides an overview of the methodologies for risk and cost monitoring for VVT and proposes a novel approach for modeling VVT strategies as decision problems. A quantitative VVT process and risk model is proposed. Due to the non-deterministic nature of risk, simulation is used to generate distributions of possible costs, schedules and risk outcomes. These distributions represent a probabilistic approach and are analyzed in relation to impact events. The model provides means to explore different VVT strategies for optimizing relevant decision parameters.
To demonstrate the proposed procedure the paper describes a case study depicting a planned avionics suite upgrade program for a fighter aircraft. Some simplified partial quantitative results are also presented.
[1]
Tyson R. Browning,et al.
Sources of Schedule Risk in Complex System Development
,
1999
.
[2]
B Dean Edwin,et al.
Cost Risk Analysis Based on Perception of the Engineering Process
,
1986
.
[3]
Avner Engel,et al.
A methodology for modeling VVT risks and costs
,
2003
.
[4]
Paul R. Muessig,et al.
Optimizing the selection of VV&A activities: a risk/benefit approach
,
1997,
WSC '97.
[5]
Dan X. Houston,et al.
COST OF SOFTWARE QUALITY: A MEANS OF PROMOTING SOFTWARE PROCESS IMPROVEMENT
,
1998
.
[6]
Tyson R. Browning,et al.
3 Sources of Performance Risk in Complex System Development 1
,
1999
.