System boundaries and assumptions
5 min read
Lesson objectives
- Apply system boundaries and assumptions to case data and justify the resulting decision and limits.
Please sign in to use all the tools
Once you sign in, the reading tools are yours: larger or smaller text, typeface, line spacing, focus mode, highlighting and bold, and notes saved to your account and gathered in My study.
A team can frame a sound question and still look only inside its organization. For foresight that boundary is often too narrow. A food distribution choice depends on customers, suppliers, carriers, regulators, energy, and weather even if one company controls the budget. A system map does not attempt to draw the entire world. It shows what could change the choice, how the parts connect, whose voice is heard, and whose experience is missing. Begin with the service to protect, then trace flows of goods, information, and money, and finally identify who controls each flow or suffers when it stops.
Suppose a regional distributor must choose between one central warehouse and two smaller ones. Its team says suppliers deliver within a day and transport between cities is stable. These are load-bearing assumptions, not permanent facts. Draw supplier → stock → vehicle → delivery point → customer. For each arrow ask: what is the unit, time window, and evidence? Supplier delay might be measured in days per shipment, vehicle outage in hours per trip, and failed access as a share of orders within a defined area. Do not compare those figures before aligning their scope and period. The map reveals that a central warehouse may save storage cost but concentrate the impact of transport disruption. Two warehouses can shorten distance but increase inventory cost. The map identifies a test; it does not choose a winner in advance.
Include social boundaries. Who still receives service during a disruption? An average delivery time can hide remote communities behind stronger urban performance. Add a representative of peripheral customers, drivers, and a small supplier to the perspectives examined. Their opinions need not be treated as equivalent to measured counts; their inclusion prevents the head office perspective from becoming the only assumption. Show disagreements openly. An operations manager might claim all deliveries take one day, while records for four recent weeks show 7 of 40 shipments arrived after two days. That is 7/40=17.5% in this sample. It raises the priority of testing the delivery assumption; it does not establish a future delay probability of 17.5%.
An assumption register makes the map revisable. Record each assumption, current evidence, the effect if it fails, a low-cost test, and a review date. For “supplier delay stays below one day,” a manager's statement without complete records is weak evidence; failure could empty the central stock; a test is to calculate arrival time for the last 40 shipments using consistent start and end points; review after a month. For “peripheral customers can reach one depot,” do not settle the issue by guessing from a map. Compare actual travel time and service hours. Test first the assumption whose failure would change the choice, not merely the one easiest to measure.
Lesson summary
For system boundaries and assumptions, inspect case data and compare exercise output with the worked denominators and assumptions.