LIVE TOOL
Value stream map (VSM)
Map a value stream from supplier to customer, current and future state: processes with their data, stock, push, FIFO and supermarket links, information flow, kaizen bursts, lead time, value-added time and PCE.
Example map · Illustrative data: an imaginary bracket line, not from a real plant, customer or supplier. Change anything, open a file or start a new map.
Map
Results
12.18 dLead time
58.7 sTakt
126.0 sValue-added time
0.019%Process cycle efficiency
Current state: lead time 12.18 working days, value-added 126.0 s, bottleneck Assembly (51.6 s against takt 58.7 s)
| Metric | Current | Future |
|---|---|---|
| Takt | 58.7 s | 58.7 s |
| Lead time (working days) | 12.18 | 3.07 |
| Stock in days | 12.17 | 3.07 |
| Value-added time | 126.0 s | 93.0 s |
| Process cycle efficiency | 0.019% | 0.056% |
| Operators | 4 | 4 |
| Bottleneck | Assembly | Weld and assemble cell |
| Process | Link | Stock (pcs) | Stock days | CT (s) | Uptime × yield | Effective CT (s) | Capacity / day |
|---|---|---|---|---|---|---|---|
| Stamping | Push | 4600 | 5.00 | 1.0 | 85.0% | 1.2 | 45900 |
| Spot welding | Push | 1100 | 1.20 | 39.0 | 100.0% | 39.0 | 1385 |
| Assembly | Push | 1600 | 1.74 | 46.0 | 89.1% | 51.6 | 1046 |
| Packing | Push | 1200 | 1.30 | 40.0 | 100.0% | 40.0 | 1350 |
Symbols on the map
Unverified: Value stream mapping symbols as commonly used, written from memory; there is no single standard set. Drawn in FactoKit's own simplified style.
- Outside source
- Supplier or customer, a factory outline with a saw-tooth roof.
- Process box
- One process through which material flows continuously; its data box lists cycle time, change-over, uptime, operators, batch and scrap.
- Inventory
- Triangle with I: stock waiting in front of a process or before shipping, in pieces and days of demand.
- Push arrow
- Thick striped arrow: material is pushed to the next process by a schedule, whether it needs it or not.
- FIFO lane
- Two lines with FIFO and a maximum: first in, first out, with a limited quantity between two processes.
- Supermarket
- Open shelves: a controlled store from which the next process pulls what it needs.
- Electronic information
- Zigzag line: orders, forecasts or schedules sent electronically.
- Manual information
- Straight thin line: information on paper or by word.
- Kaizen burst
- Jagged star: an improvement needed to reach the future state.
- Timeline
- Stock days above, cycle times below; lead time, value-added time and PCE at the end.
Download offline versionSingle HTML file · v1.5 · 561 KB · works without internet
How the value stream map is calculated
A value stream map follows one product family from the supplier to the customer: the processes, the stock between them, how material moves and how information tells each process what to make. Value stream mapping is a common practice, not a standard; the symbols and metrics here are the usual ones, drawn in FactoKit’s own simplified style.
Formulas
- Available time per day = available time per shift × shifts per day; takt = available time per day / customer demand per day.
- Stock days = pieces in stock / demand per day, for the stock in front of each process and the finished goods.
- Effective cycle time = cycle time / (uptime × (1 − scrap)); it is compared with takt. Capacity per day = the process’s own available time per day (its shifts and time, or the stream’s) / effective cycle time; the process with the lowest capacity per day is the bottleneck (the longest effective cycle time when every process works the stream’s time).
- Lead time (working days) = Σ stock days + finished goods days + Σ cycle times / available time per day.
- Value-added time = Σ cycle times; process cycle efficiency (PCE) = value-added time / (lead time × available time per day).
Worked example (the tool’s example)
- 920 pieces a day in two shifts of 27000 s: takt = 54000 / 920 = 58.7 s.
- Current state: 4600 + 1100 + 1600 + 1200 pieces in front of the four processes and 2700 finished are 5.00 + 1.20 + 1.74 + 1.30 + 2.93 = 12.17 days; the cycle times 1 + 39 + 46 + 40 = 126 s add 0.002 days: lead time 12.18 days, PCE 126 / (12.18 × 54000) = 0.019 %.
- Assembly (46 s, uptime 90 %, scrap 1 %) runs at an effective 51.6 s: the bottleneck, still under takt.
- Future state: SMED at the press cuts the batch and its stock to 1500 pieces, a weld and assembly cell pulls from a supermarket of 300, a FIFO lane of at most 150 feeds packing, and one day of finished goods is kept: lead time 3.07 days, value-added time 93 s, PCE 0.056 %.
Common mistakes
- Mapping from the ERP or the routing instead of walking the stream: the real stock and cycle times differ.
- Mapping every product at once: one map is for one product family with similar steps.
- Adding cycle times that run in parallel: two machines doing the same step are one process with half the effective cycle time per piece.
- Drawing a future state nobody can reach: every change in it needs a kaizen burst and an owner.
- Reading PCE as a target for a process: the waiting between processes is where the time goes.
Input and files
- Processes are listed in flow order; the up and down buttons move them, a click on a box in the map chooses it. Numbers may use a decimal comma or point; an empty change-over or batch is shown as a dash.
- On this website the map is saved in this browser (IndexedDB) as you work, one map at a time; Save JSON keeps it as a file that opens again here or in the offline file. The offline file stores nothing.
- Excel export: a sheet per state with takt, stock days, effective cycle times, lead time, value-added time and PCE as formulas, and a summary comparing the states. SVG and PNG export the map shown for editing; Print / PDF puts both maps, the metrics and the kaizen bursts on one A3 landscape sheet.
- The blank process data sheet (fillable PDF or Excel) is for collecting the data while walking the stream.
Questions
- What is the difference between the current and the future state map?
- The current state shows the value stream as it runs today, drawn from what is seen on the floor: stock counted, cycle times observed. The future state shows how it should run after the improvements, with the kaizen bursts marking what has to change to get there. Make the future state as a copy of the current one and change it.
- How is the lead time on the timeline calculated?
- Each stock is turned into days of customer demand (pieces / demand per day), because a piece waits on average that long before it moves on. The lead time is the sum of these days, the finished goods included, plus the cycle times of the processes expressed in working days. It is a working-day figure, not calendar days.
- What does process cycle efficiency (PCE) tell me?
- PCE = value-added time / lead time, both in working seconds. In most current-state maps it is far below 1 %: a part spends almost all its time waiting in stock. It is a way to show where time goes, not a target for one process; shorten the waiting before speeding up the processes.
- Which process is the bottleneck?
- The one with the longest effective cycle time: the cycle time divided by uptime and by the share of good parts. A process whose effective cycle time is above takt cannot keep up with the customer even when everything else is perfect; the map flags it.
Related tools
- Takt time calculator: takt and station loads in more detail.
- SMED change-over analysis: the change-over behind a batch on the map.
- Kanban sizing: the cards for a supermarket on the future state.
- Little’s law and WIP: lead time from work in process and throughput.