Vose SoftwareModelRisk example · Manufacturing
Supply-Chain Disruption
What do supplier disruptions cost us, and which protection pays? Safety stock, a second source or insurance.
From the model’s saved run
What do disruptions cost?
Doing nothing: days the plant stands still, and the contribution lost (US$m).
| Measure | Unit | 5th pct | Median | 95th pct | Mean |
|---|---|---|---|---|---|
| Days lost | days | – | 2.0 | 58.7 | 13.0 |
| Contribution lost | US$m | – | 0.49 | 14.67 | 3.25 |
| Measure | Unit | Value |
|---|---|---|
| Chance of losing any days in a year | 54.6% | |
| Chance of losing more than 30 days | 14.2% | |
| 1 year in 100 (99th percentile) | US$m | 28.48 |
Histogram of Days lost from the saved simulation.
The business question
An appliance maker buys castings and electronics from two suppliers in the same coastal region, motors from a supplier abroad and packaging locally — one source each. Every so often a supplier stops: a fire at the foundry, a chip allocation, a strike, a missed delivery.
Ten days of safety stock absorb most of these, but a long disruption stops the plant, losing about US$250,000 of contribution a day. And a typhoon or an earthquake in the coastal region would stop castings and electronics together. The board is weighing three protections: thirty days of stock for the critical parts, a second electronics supplier elsewhere, or business-interruption insurance.
- Who it is for
- Supply-chain and procurement managers, risk managers, plant and finance directors.
- Main outputs
- The plant's lost days and lost contribution in a year, average and bad years · each supplier's share · the annual cost of each protection option, including its own cost, average, 1 in 20 and 1 in 100 years.
How the model works
Three sheets after this one, read from left to right. Each one feeds the next.
Inputs sheet
The contribution lost per day stopped and today's safety stock. For each supplier and for the regional disaster, how often a disruption happens and how long it lasts; the cost and effect of each option.
Disruptions sheet
Each iteration draws the number of disruptions of each source (Poisson) and the length of each (Lognormal). Safety stock covers the first days; the rest stops the plant. The same disruptions are then replayed with more stock, with a second electronics supplier and with insurance.
Results sheet
Days lost and contribution lost in a year; each source's share; each option's total annual cost — its own cost plus the losses it leaves — on average, 1 year in 20 and 1 year in 100.
ModelRisk techniques
Find similar models
- Product
- Type
- Application
- Uses
The results in detail
Which source stops the plant?
Average days lost a year to each source, doing nothing.
| Source | Rate a year | Length days, mean | Days lost a year | Share of days lost |
|---|---|---|---|---|
| Castings (East) | 0.30 | 20 | 3.23 | 25% |
| Electronics (East) | 0.25 | 30 | 5.17 | 40% |
| Motors (West) | 0.40 | 15 | 2.57 | 20% |
| Packaging (local) | 0.50 | 5 | 0.22 | 2% |
| Regional disaster, East | 0.05 | 45 | 1.82 | 14% |
Which protection pays?
Each option's annual cost: what it costs, plus the losses it leaves (US$m).
Chart: Do nothing, More safety stock, Second electronics supplier, Business-interruption insurance.
| Option | Average | 1 in 20 95th pct | 1 in 100 99th pct | 1 in 200 99.5th pct | Lowest average |
|---|---|---|---|---|---|
| Do nothing | 3.25 | 14.67 | 28.48 | 34.69 | |
| More safety stock | 1.95 | 8.30 | 21.81 | 28.40 | ✔ lowest |
| Second electronics supplier | 2.56 | 10.33 | 19.79 | 23.98 | |
| Business-interruption insurance | 3.66 | 7.14 | 15.48 | 21.69 |
How to read these results
- Doing nothing, disruptions stop the plant in about half of all years, for 13 days on average. One year in 20 it loses about 59 days — US$15 million of contribution — and one year in 100 about US$28 million.
- Electronics cause 40% of the days lost: disruptions there are no more frequent than at the foundry, but longer and more variable. The regional disaster is rare — one year in twenty — but stops two suppliers at once for weeks: 14% of the days lost.
Thirty days of stock is the cheapest protection on average — about US$1.95 million a year with its holding cost, against US$3.25 million for doing nothing — and nearly halves the 1-in-20 loss. A second electronics supplier saves less on average but cuts the very worst years a little more.
As shipped it takes over in 7 days, before the 10 days of stock run out, so it ends every electronics stop on its own: what it leaves is the other suppliers and the regional disaster.
- Insurance costs more on average than doing nothing — its premium is above its expected payout — but it protects best against bad years: about US$7 million 1 year in 20 and US$15 million 1 year in 100. The options answer different questions: what is cheapest on average, and what the company can survive.
- Combinations are worth testing: more stock for castings and motors, a second source for electronics, insurance for the regional disaster.
See it move. Open the workbook, change an input on the Inputs sheet and run the simulation: every table and chart on this page updates in Excel.
Inputs and assumptions
Every assumption in the model, with its uncertainty and where it comes from. The figures are illustrative.
The plant 2 inputs
| Assumption | Unit | Value |
|---|---|---|
| Contribution lost per day stoppedAbout US$62m a year over 250 days. | US$m | 0.25 |
| Safety stock todayOf every purchased part. | days | 10 |
Suppliers 15 inputs
How often each source is disrupted, and for how long.
| Assumption | Unit | Value |
|---|---|---|
| Castings (East): disruptionsSingle foundry; fire or breakdown. | a year | 0.30 |
| Castings (East): average length | days | 20 |
| Castings (East): standard deviation of length | days | 15 |
| Electronics (East): disruptionsChip allocation; long and uncertain. | a year | 0.25 |
| Electronics (East): average length | days | 30 |
| Electronics (East): standard deviation of length | days | 30 |
| Motors (West): disruptionsQuality holds and strikes. | a year | 0.40 |
| Motors (West): average length | days | 15 |
| Motors (West): standard deviation of length | days | 10 |
| Packaging (local): disruptionsShort hiccups. | a year | 0.50 |
| Packaging (local): average length | days | 5 |
| Packaging (local): standard deviation of length | days | 4 |
| Regional disaster, East: disruptionsA typhoon or earthquake stops castings and electronics together. | a year | 0.05 |
| Regional disaster, East: average length | days | 45 |
| Regional disaster, East: standard deviation of length | days | 30 |
Options 7 inputs
| Assumption | Unit | Value |
|---|---|---|
| 1 More stock: safety stock of castings, electronics, motors | days | 30 |
| 1 More stock: cost of holding it20 days × US$150,000 of parts a day × 25% a year. | US$m a year | 0.75 |
| 2 Second source: days to take over the electronicsA qualified supplier in another region. | days | 7 |
| 2 Second source: price premium3% on US$20m of electronics. | US$m a year | 0.60 |
| 3 Insurance: waiting periodDays stopped before cover starts, per event. | days | 10 |
| 3 Insurance: limit | US$m a year | 15.0 |
| 3 Insurance: premiumThe insurer's expected payout plus its margin. | US$m a year | 2.00 |
Units and conventions
- Money
- US$ million (US$m) a year.
- Days lost
- Days the plant stands still: a disruption's length beyond the safety stock.
- Annual cost
- Contribution lost plus the option's own cost, less any insurance recovery.
- 1 in 20, 1 in 100, 1 in 200
- The 95th, 99th and 99.5th percentiles of the annual cost.
Key assumptions and limitations
- Disruptions
- Independent between suppliers, apart from the regional disaster; at most five per source a year.
- Overlap
- Days lost to different disruptions are added: overlaps are rare.
- Recovery
- Production restarts as soon as parts arrive; lost sales are not recovered later.
- Data
- The plant, the suppliers and all figures are invented for this example.
How to use it
- Enter your daily contribution and safety stock.
- Enter how often each supplier is disrupted and for how long — and any common causes.
- Run a simulation from the ModelRisk ribbon (10,000 samples with a fixed seed are set).
- Compare the options on average cost and on bad years; they answer different questions.
ModelRisk functions used
- VosePoisson
- The number of disruptions of a source in a year.
- VoseLognormal
- The length of a disruption, by its mean and standard deviation.
- VoseSimMean, VoseSimPercentile, VoseSimProbability
- Statistics of the simulation on the Results sheet.
Further reading
- Monte Carlo simulation explained (Vose guide)
- Sheffi, Y. (2005). The Resilient Enterprise. MIT Press.
- Vose, D. (2008). Risk Analysis: A Quantitative Guide, 3rd edition. Wiley — risk events.
Download the workbook and run it yourself. It opens in Excel with ModelRisk. The free trial runs every example model.
Related example models
- ModelRisk★★★★★Capacity Expansion DecisionHow many presses should we add?Sixty months of seasonal, uncertain demand against four capacity options with overtime, subcontracting and lost sales.
- ModelRiskOptimizer★★★★★Spare Parts and Maintenance Strategy Chosen by the OptimizerWhich replacement age and how many spare kits cost least while keeping the downtime promised to a customer?Bearing lives simulated on every spindle; the replacement age and spare kits chosen together by the Optimizer within a limit on downtime.
- ModelRiskOptimizer★★★★★Inventory Policy Chosen by the OptimizerWhich reorder point and order quantity cost least while keeping the service the customers were promised?Demand over a random lead time as an aggregate; the reorder point and order quantity chosen together by the Optimizer for a service target.
An illustrative example: the figures are invented for teaching and do not describe a real company, project or asset. Last updated 9 October 2026.