5.5Risk management & supply chain resilience
The essentials in 30 seconds
- Map probability × impact: list risks and focus effort on 3 to 5 major ones.
- Hunt single points: with no pre-qualified plan B, shock becomes shutdown.
- Prepare the response: signals, 3 actions, owner and already-tested alternative.
Mini-case thread — Nadia pre-qualifies her plan B
Nadia depends on a single supplier for a critical imported ingredient at her agrifood SME in Algiers: if the route stops, production stops. This section shows how she rates her risks, qualifies a second source in advance and writes a one-page plan tested on table.
Last episode of cases 5.1-5.4 — Nadia closes her steering with resilience.
1. Map: probability × impact
Recent shocks (pandemic, blockages, shortages, geopolitics, climate) showed the fragility of cost-only optimised chains. Question: «will we stand firm?».
List the risks: suppliers (financial failure 2.1, capacity, incident, dependency), logistics (sea route, freight), geo/geopolitical (country concentration), market (materials, currencies), internal (key person 5.2, single site, IS), climate.
Rate probability × impact (1-3): the matrix prioritises 3-5 major risks. Hunt single points of failure: single supplier/route/person/site — where shock becomes shutdown.
2. Treat: the supply chain palette
Avoid/reduce: diversification (multi-zone dual sourcing, routes/modes 4.3, nearshoring — bringing closer at sometimes higher price but winning TCO+risk), buffers (costed strategic stock: holding vs probabilised stop cost — 5.3), secured contracts (capacity, priority 2.3).
Visibility: map tiers 2-3 on critical chains (0.2), monitoring (supplier health, market, source zones) and relationships that circulate early warning (2.5).
Transfer: insurance (goods 4.3, business interruption, customer failure), currency hedging (bank, varying by country). Consciously accept some costed risks — what matters is having decided it.
3. Prepare the response: the continuity plan
For each major risk: warning signals, 3 first actions, owner, pre-qualified alternative (audited supplier B, tested route B).
In a crisis, those with a plan execute while others discover. Table-test one scenario (critical supplier down tomorrow: what do you do hour by hour?).
Resilience costs a little every day and saves everything when the day comes: the well-designed chain is not the cheapest in calm weather, it is the one that stands firm in rough weather.
🔎 Concrete example — The single-sourced component
Critical distant single-sourced part: major risk. Treatment: qualification of a second source at 30 % in a different zone + 3-week strategic stock costed vs stop cost + monthly financial-health monitoring.
Plan: signal = OTIF <90 % two months; D1 action = switch 30 % → 100 % to already-tested source B; owner = purchasing.
Lesson: the switch is immediate because pre-qualified — otherwise it takes months when you need it.
Section vocabulary
- Risk mapping
- Inventory of risks (suppliers, logistics, geopolitical, market, internal, climate) rated by probability × impact; focuses action on the 3-5 major risks.
Simple example: Nadia rates each risk 1 to 3 and keeps three major ones to handle.
What it is not: listing twenty risks with no rating and no priority, all "important".
Reflex: what are my 3 major risks, written and rated?
- Single point of failure
- Element with no alternative whose failure stops everything: single supplier of a critical item, single route, single site, single person — the priority treatment target.
Simple example: Nadia spots her single-sourced ingredient with no qualified alternative.
What it is not: believing a second name on a list, never tested, is a plan B.
Reflex: what happens if this item stops tomorrow?
- Double sourcing
- Two qualified sources (ideally in different zones) for critical items; the redundancy turning a failure into a simple switch.
Simple example: Nadia qualifies a second source in a different zone at 30% before the crisis.
What it is not: adding a second supplier on rupture day, with no test or audit.
Reflex: is my second source qualified and already tested?
- Strategic stock
- Voluntary buffer on vital items, sized by calculation: holding cost vs (probabilised) cost of the stop avoided — costed insurance, not a return to overstock.
Simple example: Nadia costs three weeks of coverage against the cost of a line stop.
What it is not: inflating all inventories "just in case" with no calculation or review.
Reflex: is my inventory insurance costed against the stop avoided?
- Continuity plan (BCP)
- Prepared response per major risk: warning signals, first actions, owner, PRE-QUALIFIED alternatives — in a crisis, you execute instead of discovering.
Simple example: Nadia writes: OTIF signal, D1 switch to source B, purchasing owner.
What it is not: a filed plan never tested, with no signals and no named owner.
Reflex: does my plan fit on one page and has it been tested?
In practice — Map your risks and build your resilience plan
- List your risks by family (suppliers, logistics, geo, market, internal, climate) and rate each by probability × impact (1-3): your matrix reveals the 3-5 majors.
- Hunt single points: which supplier/route/person/system has no alternative? What happens if it stops tomorrow and for how long?
- For each major risk, choose the treatment: diversify (2nd source), buffer (costed strategic stock), transfer (insurance), or consciously accept.
- Write a one-page BCP per major risk (signals, 3 actions, owner, pre-qualified alternative) and table-test one.
Ready-to-say script — test the plan on table (1 min)
"Imagine our critical supplier stops tomorrow: what do we do hour by hour? Who calls the already-qualified source B, who handles customers, who tracks inventory? We note gaps and fix the one-page plan before the real crisis."
Say in team exercise, then update signals, actions and owner.
Copy-paste template — one-page continuity plan
MAJOR RISK: _______________
WARNING SIGNALS: _______________
3 FIRST ACTIONS (D1): _______________
OWNER: _______________
PRE-QUALIFIED ALTERNATIVE: _______________
TABLE TEST (date + gaps found): _______________Small business / tight budget
One page per major risk is enough: Nadia tests a dry scenario with no costly tool.
Algeria SME
Pre-qualify route and source B in advance: immediate switch avoids months of qualification in crisis.
Agrifood import
Cover the critical ingredient with dual sourcing and costed inventory: line stop costs more than insurance.
Key person and single site
Duplicate skill and document: the single person or site is also a point of failure.
✓ Best practices
- Qualify the second source BEFORE you need it: qualification takes months, crisis takes hours.
- Cost the stock insurance: holding vs probability of stop × cost of stop — the decision becomes objective.
- Circulate the warning: a partner warns, a pressured supplier stays silent.
✗ Common pitfalls
- Believing risk stops at your direct suppliers: tiers 2-3 often decide your availability.
- Having an unqualified alternative: on D-day, it is not an alternative.
- Plan filed in a drawer untested: table testing reveals gaps before the real crisis.
Key takeaways
- Map probability × impact and hunt SINGLE POINTS OF FAILURE.
- Diversify (multi-zone, routes, nearshoring), buffer (costed stock), visibility (tiers 2-3, monitoring), transfer (insurance), consciously accept.
- A simple BCP per major risk (signals, actions, owner, pre-qualified): you execute instead of discovering.
- Resilience costs little and saves everything when the day comes; the well-designed chain stands firm in rough weather.
48h challenge — apply tomorrow
List your single points (supplier, route, person, site with no alternative) and rate probability × impact to isolate 3 major risks. For the first, write the one-page plan on the template above.
- My 3 major risks are rated.
- My single points are listed with shutdown time.
- My first one-page plan is written.
Frequently asked questions
How to cost a strategic stock?
Compare annual holding cost (15-30 % of value) vs probabilised cost of the stop avoided (annual probability × cost of a stop: lost sales, penalties, image). If insurance is cheaper than expected loss, it is justified — and sized (weeks of coverage) to exposure.
Should everything be duplicated in dual sourcing?
No: focus redundancy on critical items (shutdown if stockout) and families with geopolitical/logistics risk. On substitutable routine items, assumed single sourcing with periodic re-tendering stays more efficient.
What is a single point of failure?