SLA Configuration Tool

Explore and tune your SLA posture. Only change inputs — the engine handles the math and relationships.

How the SLA Math Works

Variables:

  • E = Elapsed Days
  • A = Allowed SLA Days
  • D = Expected Duration
  • θ = Grace Window
  • W = Weight
  • ΣDᵢ = Sum of Delay Penalties

Equations (natural language style):

  • Time Risk = E / A
  • Stage Risk = E / (D + θ) × W × Severity
  • Delay Risk = ΣDᵢ
  • Total SLA Score = Time Risk + Stage Risk + Delay Risk

This tool is for exploration, not policy. Use it to see how your choices behave before committing to configuration values.

How to Use This Tool

  • Set duration logic values (Customer, Default, Failsafe SLA).
  • Configure stage: expected duration, grace window, weight, and severity.
  • Enter elapsed days for a scenario.
  • Toggle delay vectors on/off and adjust penalties.
  • Watch Time, Stage, Delay, and Total SLA scores update live.
  • Use the color status as a quick posture indicator.

Once you like how it behaves, document your chosen values and implement them in your SLA configuration store (e.g., WF‑SLA‑Stage‑Config and WF‑SLA‑Duration‑Config).

Duration Logic Inputs Editable

These values determine the total allowed SLA days for a ticket.

Logic: use Customer SLA if > 0, otherwise Default SLA if > 0, otherwise Failsafe.

Stage Configuration Editable

Configure how this stage behaves in the SLA engine.

Stage Risk uses grace as a buffer: E / (D + θ) × W × Severity.

Delay Vectors Editable

Select delays that occurred and tune their penalties. Delay risk is additive.

No tech available
Parts unavailable
Parts backordered
Customer‑requested pause

Delay Risk = Σ of penalties for all selected delay vectors.

Results

Live SLA posture based on your current inputs.

Allowed SLA Days (A)
–
Resolved from duration logic
Time Risk
–
E / A
Stage Risk
–
E / (D + θ) × W × Severity
Delay Risk
–
Σ delay penalties
Total SLA Score
–
Time Risk + Stage Risk + Delay Risk

Thresholds (example): < 1.0 = Green, 1.0–1.2 = Yellow, > 1.2 = Red. Tune these in policy, not here.