When a supplier outage or a sudden demand spike threatens production, the conversation quickly turns to resilience. Over the years I’ve helped OEMs and tier‑1 suppliers translate that fuzzy idea into numbers that executives can act on. In this piece I’ll walk you through practical, evidence‑based ways to quantify the resilience benefits of two classic tactics: dual sourcing and buffer inventory. I’ll show the metrics that matter, simple models you can run quickly, and how to combine probabilistic disruption analysis with cost accounting so you can make a defensible investment decision.
Why quantify resilience?
Resilience is often framed as a qualitative strategic benefit. But plant managers and CFOs need dollars, days, or percent‑points of service improvement to approve changes. Quantification lets you:
Key metrics to use
Start with a small set of metrics that capture both operational and financial impact. I recommend tracking these:
Simple model: single supplier with buffer inventory
A practical first step is a deterministic plus probabilistic lead‑time model. Assume:
You can compute safety stock via the classic formula:
Safety stock = z × sqrt( (LT × σd^2) + (D^2 × σLT^2) )
The choice of z maps to desired cycle service level. Convert safety stock to ICC:
ICC = Safety stock × unit cost × carrying rate
Estimate ESC from the probability of stockout multiplied by expected shortage per stockout and unit shortage cost. For many practical cases, a Normal demand assumption with loss function gives a good approximation for ESC.
Simple model: dual sourcing without buffers
Dual sourcing adds a second procurement path. Key parameters:
A simple way to model is to treat supply availability as a Bernoulli process each ordering cycle. If you source a fraction α from supplier A and (1−α) from B, the probability both suppliers fail in the same period is approximately SDP1 × SDP2 (if independent). The expected shortage probability is reduced nonlinearly as α is diversified.
Putting it together: expected cost formulation
For decision making I use a Total Expected Cost per period (TEC) that sums inventory and disruption impacts:
| Component | Formula / Description |
| Inventory carrying cost | ICC = (Safety stock) × unit cost × carrying rate |
| Expected shortage cost | ESC = P(stockout) × Avg shortage quantity × unit shortage cost |
| Procurement cost | PC = α×p1×D + (1−α)×p2×D + dual‑sourcing premiums |
| Operational overhead | OS = contract management + quality management + logistics for second supplier |
| Total Expected Cost | TEC = ICC + ESC + PC + OS |
Minimize TEC over decision variables (safety stock, α, order quantities) subject to service constraints. For many plants a grid search or simple optimization (e.g., Nelder‑Mead) is sufficient and interpretable for stakeholders.
Using probability and simulation: Monte Carlo for realism
When demand and lead times are non‑normal or suppliers have correlated disruptions (common cause events), Monte Carlo simulation is the pragmatic way to quantify resilience benefits. I typically simulate thousands of periods for scenarios like:
From the simulation we extract:
Example: a quick back‑of‑envelope
Suppose unit cost £100, D=1,000 units/month, carrying rate 25% annual, target CSL 95% with safety stock 500 units. ICC = 500×100×0.25/12 ≈ £1,042 per month. If unserved unit shortage cost (lost margin & rush freight) = £300 and probability of stockout per month = 2%, ESC = 0.02×average shortage (say 200 units)×300 ≈ £1,200/month. TEC (ignoring procurement and overhead) ~ £2,242/month.
Now add a second supplier that halves the joint disruption probability and increases unit cost by £2. If dual sourcing reduces P(stockout) to 0.5% and contractual overhead = £500/month, ESC becomes 0.005×200×300=£300. New TEC ≈ ICC + ESC + extra procurement cost + overhead. If procurement cost increase is small compared with ESC reduction, dual sourcing wins. This shows why you should quantify using your own numbers rather than instincts.
Practical tips for implementation
How I present results to stakeholders
I avoid heavy math in the boardroom. I prepare a short dashboard: TEC per month (or year) for each option, expected days of downtime per year, likelihood of missing customer commitments, and payback period for the incremental cost of dual sourcing or additional inventory. A single chart showing TEC vs. service level often closes the discussion faster than pages of equations.
If you want, I can share a simple Excel/Google Sheets template that runs the basic TEC comparison and a Monte Carlo prototype you can adapt to your parts and suppliers. I frequently use this approach in pilot projects — it’s a quick, transparent way to move from “feels right” to “ROI‑backed” resilience planning.
