Skip to main content

User Story: Risk Score Calculation (AX02US05)

Step 1: Normalize the Input Variables

The normalization for Criticality and Condition remains the same. The key improvement is in how we normalize the asset's age.

  1. Normalized Criticality (C_norm):
    • C_norm = Criticality / 5
  2. Normalized Condition (Cond_norm):
    • Assuming a 1-5 scale where 5 is the worst condition.
    • Cond_norm = Condition Score / 5
  3. Normalized Age (Age_norm) - Revised & More Accurate:
    • This component should represent the fraction of the asset's total life that has been consumed.
    • Inputs: Total Useful Life (TUL) and Remaining Age.
    • Consumed Life = Total Useful Life - Remaining Age
    • Age_norm = Consumed Life / Total Useful Life
    • Example: An asset has a TUL of 40 years and a remaining age of 10 years.
    • Consumed Life = 40 - 10 = 30 years
    • Age_norm = 30 / 40 = 0.75. This correctly shows the asset is 75% through its planned life.

Step 2: Calculate the Probability of Failure (PoF) - Now with MTBF

Now we have three distinct factors influencing the probability of failure: its current Condition, its Age (as a fraction of total life), and its historical reliability (MTBF). We can combine these into a single PoF score.

First, let's normalize the MTBF. A lower MTBF means more frequent failures and thus a higher probability of failure. So we need to invert its effect.

  • Normalized MTBF (MTBF_norm):
    • We need a reference point. Let's call it Max_MTBF. This would be the MTBF of the most reliable asset type in your system. If you don't have a maximum, you can use a theoretical "ideal" MTBF for a new, perfectly performing asset.
    • Consider the Max_MTBF as the number of houses between preventive schedule of the asset
    • MTBF_norm = 1 - (MTBF / Max_MTBF)
    • This formula correctly maps a high MTBF to a low score and a low MTBF to a high score.
    • Example: An asset has an MTBF of 15 years, and the best-in-class asset (Max_MTBF) has an MTBF of 25 years.
    • MTBF_norm = 1 - (15 / 25) = 1 - 0.6 = 0.4.

Now, we combine the three normalized factors using a weighted average to calculate PoF. This allows you to tune the influence of each component.

  • Weights:
    • Weight_Condition (W_c) = 0.5 (50% influence) - Often the most important factor.
    • Weight_Age (W_a) = 0.3 (30% influence)
    • Weight_MTBF (W_m) = 0.2 (20% influence) (Note: The sum of weights should be 1. These should be configurable in your software.)
  • PoF Formula:
    PoF = (Cond_norm * W_c) + (Age_norm * W_a) + (MTBF_norm * W_m)
  • Example PoF Calculation:
    • Cond_norm = 0.6 (for a condition score of 3)
    • Age_norm = 0.75 (from our age example)
    • MTBF_norm = 0.4 (from our MTBF example)
    • PoF = (0.6 * 0.5) + (0.75 * 0.3) + (0.4 * 0.2)
    • PoF = 0.30 + 0.225 + 0.08 = 0.605

Step 3: Calculate the Final Risk Score (out of 100)

This step remains the same: multiply the normalized criticality by the comprehensive PoF and scale to 100.

Risk Score = C_norm * PoF * 100

  • Final Example Calculation:
    • C_norm = 0.8 (for a criticality of 4)
    • PoF = 0.605
    • Risk Score = 0.8 * 0.605 * 100 = 48.4

The final risk score for this asset would be 48.4 out of 100.


Back to top