دانلود رایگان مقاله انگلیسی رویکرد سیستم ها به اعتبار سنجی تحلیلی ریسک برای مدیریت ریسک - الزویر 2017

abstract

This paper presents a logical structure to address the topic of this special issue: Risk Analysis Validation and Trust in Risk Management. We do that by presenting a systems approach that links all four of those concepts. The underlying logic: Validation should test how effectively a risk analysis supports actual, real-world implemented risk management. Our approach is based on a flowchart linking all of the elements from inputs through risk analysis, risk reporting and transparency, then how that reporting-transparency support the risk management decision making process and associated third party and stakeholder reviews (formal or informal), which in turn determine the trust and acceptance necessary for the real-world implementation of risk management actions. We take that flowchart and identify within it sixteen critical elements, then specify a validation test for each of those elements. Validation, then, consists of subjecting the risk analysis to those sixteen tests. Those tests, together, test the risk analysis for how effectively it supports implemented risk management. Another key feature: We divide the flowchart into Analysts’ Domain, Users’ Domain, and Analysis Community Domain. The Analysts’ Domain is where the risk analysts work, then the Users’ Domain stands between their work and implementation. The Analysis Community Domain is comprised of the communities of risk analysts and commissioners of risk analyses. Those two communities are where we would, as part of building our systems approach to risk analysis validation, build a ‘‘Culture of Analysis Quality,” where the sixteen validation tests would be enforced by both of those communities.

7. Summary and conclusion

We began this paper presenting the logical sequence upon which our reasoning is based: Validation of a risk analysis should be based on how well the risk analysis supports risk management, and assessing how well the risk analysis supports risk management should include considering how well it supports the actual, real-world decision making process, and the roles trust, third party review and stakeholder review and acceptance may play in that. Then we reworded that logic: Validation for risk management must be considered from a systems perspective, and presented that system in Fig. 1. That figure, in turn, provided the structure for a series of eight functions of risk analysis, and within those functions sixteen tests of validity, each test worded as a question. Each test is paired with a discussion of the shortfalls associated with failing that test. That Fig. 1 is divided into three domains: Analysts’ Domain, Users’ Domain and Analysis Community Domain. The Analysts’ Domain is the one domain implicitly assumed in any discussion of risk analysis validation we have seen. We add to that perspective the fact that there is a Users’ Domain separating any risk analysis from the final consequences and residual risk, which is the only thing that actually matters. That Users’ Domain includes the risk management decision process, desired implementation of risk management actions, and the actual risk management actions achieved based on the acceptance or denial of those actions, where that acceptance/denial is a result of implementers and stakeholder acceptance, which in turn is a result of the risk analysis report and its transparency and documentation. Underlying both the Analysts’ and Users’ Domains is the Analysis Community Domain, which could provide a ‘‘Culture of Analysis Quality” where the sixteen validation tests listed here would be enforced by both risk analysts and commissioners of risk analyses. Though we end by admitting that it is quite challenging to identify how to create that culture.

In conclusion, we have here presented a very broad, insightful systems perspective on the matter of how to establish risk analysis validity, including sixteen concrete tests of validity.

The authors are grateful to the four anonymous reviewers for their useful comments and suggestions to the original version of this paper. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.