How to separate ‘size’ cases from ‘consequence’ cases

What to read first: Worst case analysis: When, why, and how

During risk identification, you identify potential events and their consequences.

You may be very unsure of the consequences of a given event type. The main post recommended analysing the reasonable worst-case consequences and the likelihood of those worst-case consequences following.

Consider the other example mentioned in the introduction: power failures in the enterprise data centre. Assume that there are no risk controls in place, such as a back-up data centre or automatic fail-over.

A power failure can be trivially brief. Then again, it might last for weeks. The business consequences can range from nothing to total enterprise failure—or something worse, for critical infrastructure.

In assessing the risk from power failure in the data centre, you will need to understand a range of different cases for the effects on the enterprise. That will be a difficult and complicated thing to understand.

There is a way to simplify the analysis of cases. You can recognise that the consequences of a such a power failure vary at two levels. First, the power failure can last from a second or for weeks and months. Second, a power failure of any given duration can have a wide range of effects on enterprise objectives. Some kinds of effect, such as loss of market share, may or may not follow at all. Other kinds of effect, such as extended customer delivery times, may vary widely in size.

From that thought, you can identify cases in two dimensions: the power failure duration, and the enterprise effect of a given failure duration.

As a simpler example, consider the enterprise risk from uncertain sales of a product launched in a new market. The sales volume can deviate from expectations by varying amounts, as illustrated in the main post. Beyond that, a deviation within a given range has an uncertain effect on enterprise profitability. (In the main post, that relationship was assumed to be fixed. This page will recognise its uncertainty.)

Cases Total Acceptance
Extreme RWC Fair Most likely Fav-our-able Excel-lent
Profitability consequence levels Likelihoods 5% 15% 35% 30% 13% 2% 100%
Worst imaginable The enterprise becomes insolvent and has debt that cannot be settled… Conditional 50% Max
1%
Absolute 2.5% 3%
Bad Enterprise is making losses that cannot be sustained. Conditional 30% 25% Max
5%
Absolute 1.5% 3.8% 5%
Disappoint-ing Enterprise is making a loss. There may be strategic reasons to persist with it… Conditional 20% 60% 30% Max
20%
Absolute 1% 9% 10% 21%
Qualified success Enterprise is just breaking even, stakeholders looking elsewhere… Conditional 15% 40% 15% Max
30%
Absolute 2.3% 14% 4.5% 21%
Planned (success) Reasonable return on investment, justifying continuation of the activity. Conditional 30% 70% 20% N/A
Absolute 10% 21% 2.6% 34%
Good High return on investment, generating the potential for expansion. Conditional 15% 60% 20% Min
20%
Absolute 4.5% 7.8% 0.4% 13%
Excellent Very high return on investment, allowing the enterprise to move up… Conditional 20% 80% Min
1%
Absolute 2.6% 1.6% 4%
TOTALS Conditional 100% 100% 100% 100% 100% 100%
Absolute 5% 15% 35% 30% 13% 2% 100%

This table shows the deviation cases for sales volume across the top. It also shows potential enterprise consequences from each deviation, downward. The deviation cases all have their own likelihood of occurring. For any deviation case, two or three enterprise outcome consequences are possible, with conditional likelihood totalling 100%.

Each combination of case and enterprise consequence has a conditional likelihood, the likelihood of the consequence given the case. Each combination also has an absolute likelihood, the likelihood of both the case occurring and the designated consequence following. The absolute likelihood is the conditional likelihood multiplied by the case likelihood.

The likelihood of any enterprise consequence is the total of the absolute likelihoods across each consequence row. The enterprise may have set acceptable likelihoods for each consequence outcome, which determines acceptability of the risks arising from the activity as proposed. A single instance of ‘reject’ for just one risk case would mean that the activity as a whole presents unacceptable risk. In this example, there are five separate instances of ‘reject’.

In the example, profitability is at unacceptable risk from deviations in sales volume. It is at further risk from other sources, such as cost of product, salaries and expenses, interest rates, working capital availability. These sources will add to the risks totalled for deviations in sales volume, so that even if the risk from uncertain sales volume had been acceptable, the risk from selling in the new market might not have been after risk is totalled across all sources. Looking at risk acceptability in this totalled way is an innovation, recommended by the Clear Lines for simple risk applications.

In assessing risks to profitability, a real enterprise would not simply sum the risks, because sales, costs, salaries, interest rates etc. all combine to affect profit. The model recognising these interactions would be better represented as a cause and effect diagram with explicit AND and OR junctions.


Parent articles

Worst case analysis: When, why, and how

What is the reasonable worst case? Consequence and likelihood for the reasonable worst case Acceptability of the reasonable worst case Acceptability of other cases Question for risk experts

Index to the topic Risk in work unit business planning

Leave a Reply

Your email address will not be published.

Comments are moderated from a sea of spam, so may not be published immediately. Email contact may get a quicker response.