Learning objectives
- Decompose a supply chain into stages, nodes, and links
- Distinguish push from pull and where the decoupling point sits
- Read a simple value stream map and identify waiting, transport, and processing time
Stages, nodes, and links
Every supply chain can be drawn as stages (suppliers, manufacturing, distribution, customer), nodes (specific facilities, such as a plant or warehouse), and links (the lanes that connect them). A useful first exercise is to draw the chain for one product family, not for the whole company, because product families often have very different footprints. A consumer goods firm might source molded plastic components from two vendors, assemble at a single plant, store finished goods in two regional warehouses, and ship to retail via both truckload and less-than-truckload. Once drawn, mark on the diagram which links are single-sourced and which lanes are constrained (only one carrier, one port, one border crossing). Mark single points of failure at both facilities and transport links.
Push versus pull and the decoupling point
In a push system, production is scheduled against a forecast and finished goods enter inventory before any customer order exists. In a pull system, each step is triggered by actual demand from the next step downstream. Customer-order-driven production needs enough capacity and time to meet the promised delivery date. Most firms operate a hybrid in which some stages push and others pull. The decoupling point is the boundary between push and pull: upstream of it, the firm forecasts and builds to stock; downstream of it, the firm assembles or finishes to order. Moving the decoupling point upstream leaves more work to complete after the customer orders, which can reduce finished-goods stock but lengthen the wait. Moving it downstream puts more work into forecast-driven stock and can shorten delivery time. The right answer depends on demand variability, product cost, and how much customization the customer actually values.
Reading a value stream
A value stream map records for each step three numbers: processing time (when the item is actually being worked on), queue or wait time (when it is sitting between steps), and lead time (the sum). In many consumer goods chains, processing time across the entire network is a small fraction of total lead time, because material spends most of its life waiting: at a supplier, in a receiving dock, in a queue for a machine, in a finished goods warehouse, on a truck. Improving only processing time is therefore usually a low-leverage move; the bigger gains live in reducing queues and consolidating shipments. The map also makes it obvious where information hand-offs cause delay, which usually shows up as long wait times at the receiving dock of the next step.
Worked example
Problem
Map one order for a fictional product, classifying each step as processing, movement, or queue. Supplier processing 2 days, supplier queue 6 days, inbound transit 4 days, receiving inspection 1 day, queue before production 3 days, production processing 2 days, queue before packing 1 day, packing 1 day, finished goods warehouse wait 5 days, outbound transit 3 days. Treat transit as movement rather than processing. Compute (a) total processing time, (b) total movement time, (c) total wait or queue time, (d) total lead time, and (e) the processing-to-lead-time ratio.
Step by step
- Classify the ten steps. Processing: supplier processing 2, receiving inspection 1, production processing 2, packing 1. Movement: inbound transit 4, outbound transit 3. Queue: supplier queue 6, queue before production 3, queue before packing 1, finished goods wait 5.
- Total processing time = 2 + 1 + 2 + 1 = 6 days.
- Total movement time = 4 + 3 = 7 days.
- Total queue time = 6 + 3 + 1 + 5 = 15 days.
- Total lead time = 6 + 7 + 15 = 28 days, which matches the sum of the ten original step durations.
- Processing-to-lead-time ratio = 6 / 28 = 0.2143 = 21.4%.
Answer. Processing time = 6 days. Wait or queue time = 15 days. Movement = 7 days. Total lead time = 28 days. Processing-to-lead-time ratio = 21.4%, meaning roughly four-fifths of the order's life is non-value-adding wait or movement.
Practice
Work each question before opening the solution.
-
If a project cuts queue time in half at every queue without changing processing or movement, what is the new total lead time?
Show solution for question 1
New queue time = 15 / 2 = 7.5 days. New lead time = 6 + 7 + 7.5 = 20.5 days, a reduction of 7.5 days (about 27%).
-
A firm considers moving its decoupling point from finished goods (build to stock) to a later stage where only packaging is done to order. State one benefit and one cost of this shift.
Show solution for question 2
Benefit: lower finished goods inventory and less obsolescence risk because only generic units are built ahead of demand. Cost: customer lead time increases by the packaging cycle, which can hurt service unless the firm has reliable, fast final-stage capacity.
-
Why does a single-source lane (one port, one carrier, one border) usually deserve more attention than a single-source supplier in supply chain design?
Show solution for question 3
A lane failure halts every supplier using it, so the blast radius is much larger. A single-source supplier affects only the parts it supplies, and the firm can usually dual-source the part itself, while diversifying a lane often requires new infrastructure or contracts.