Extracted from Decision Analysis — A Primer:  How to Use DecisionSpeak™ and Question Charts (Q-Charts™), by Ronald G. Ross, 2013, available (free) on http://www.brsolutions.com/IPSpeakPrimers

A simulated decision essentially asks what would be the outcome if a decision about a particular matter were made at a particular point in time.

A simulated decision does not decide the matter "for real"; it simply reveals what the results would have been if the decision had been made then.  These results provide feedback that can then be used to improve results from the future "for real" decision.  Tasks calling for simulated decisions are valid and valuable in business processes.

Examples:

• Submitting a health insurance claim for a proposed procedure to determine what is covered and for how much rather than submitting the claim "for real".  Result:  A more informed decision about treatment of the patient can subsequently be made.
  
  
• Submitting a request for a product part to ascertain current pricing and availability rather than submitting the request "for real".  Result:  An optimal decision about sourcing or repairing the part can subsequently be made.

Analysis of Simulated vs. "For Real" Decisions

Decision Analysis can be undertaken for both simulated decisions and "for real" decisions.  In all cases, however, meaningful analysis can be conducted only where the decision logic is under the business jurisdiction of the organization itself.  For example, it would probably not be meaningful for:

• A health clinic to use decision analysis to understand the decision logic of an insurance company.
  
  
• A manufacturing or engineering company to use decision analysis to understand the decision logic of an external parts supplier.

On the other hand, decision logic that must be interpreted from regulatory bodies and similar governing bodies is under business jurisdiction.  Decision analysis can be quite useful in such circumstances.

DecisionSpeak Definitions decision:  a determination requiring know-how or expertise; the resolving of a question by identifying some correct or optimal choice decision rule:  a business rule that guides the making of an operational business decision; specifically, a business rule that links a case to some appropriate outcome decision logic:  the set of all decision rules for cases in scope decision analysis:  identifying and analyzing some key question arising in day-to-day business activity and capturing the decision logic used to answer the question

To continue, let's assume some targeted decision logic is under business jurisdiction of an organization.  Especially important then is recognizing situations in which that decision logic is essentially the same for both a simulated and "for real" decision.  Examples:

1. An insurance company uses the same decision logic both to:
   
   
   • answer queries about coverage for particular treatments.
     
     
   • adjudicate actual claims.
     
     
2. A manufacturing company uses the same decision logic to assess pricing and availability for parts both in:
   
   
   • composing bids for work.
     
     
   • conducting the actual work itself.

In situations like these, decision analysis needs to be performed just once since the same decision logic supports both decisions.  Unification of the analysis saves significant resources and ensures consistency in the eventual decisions.

Can the Decision Logic Change?

Unification of the analysis does not imply that the decision logic must remain completely static for any given case between every simulated and "for real" decision.

Often feedback from a simulated decision can lead directly to insights about potential improvement in the decision logic before the subsequent "for-real" decision is made.  Such adjustments are certainly possible and frequently desirable.

For many business problems the volatility in the decision logic between simulated and "for real" decisions for given cases can become quite high.  At what threshold does decision analysis cease to be useful?

The DecisionSpeak rule of thumb:  Decision analysis in the style of DecisionSpeak is fundamentally based on words (semantics) and explicit rules.  It will always remain applicable for decision analysis as long as you need to know why you get the results you get in words.

Sometimes the path and pace of simulation and change-in-the-rules becomes quite rapid, experimental, or personal.

At the point you no longer need to trace and communicate the 'why' in words, then statistics or other forms of non-verbal modeling will probably prove better suited to the problem.

# # #