What is a Value Stream?
A typical definition of VSM reads something like “a diagram of every step involved in the material and information flows needed to bring a product from order to delivery." This is somewhat misleading: if it were just about representing steps in a process, a process flowchart would be sufficient.
Here’s what's different about VSMs:
- It tracks information flows. Not just any information, but the information that triggers activity at a process step.
- It captures the location and quantity of material in the system
- It measures lead time at every step of the process and separates the processing (value-add) and the waiting (inventory) of the material, and sometimes, also the transporting of material. This lead time is visualized by a timeline underneath each process
- It represents methods by describing their outcomes: cycle times, changeover times, lot sizes, quality metrics
- It indicates the operator resources needed to run the system
- It indicates and quantifies delays related to the supply chain
Veryable's definition: “A value stream is a small, self-contained system that can transform inputs into something of value to a customer.”
Procedures for Value Stream Mapping
Step 1: Define the value stream
- It is not just a process, but a family of things that we deliver to our customer. In practice, it’s that family (e.g. “small ball valves”) which gives its name to the value stream.
- It’s OK to start with some uncertainty as to exactly what to include in the value stream and pick a representative high-runner product that you can observe being produced, just to get started.
- Build the VSM as team of 3-10 key stakeholders. Start by drawing the process sequence as a “skeleton”.
Step 2: Show the inventory quantities and locations for the materials related to your product family
- Important: the unit of measure is “equivalent finished unit”, for if RM is 7,000 pounds of steel, and the BOM indicates it takes 0.34 pounds to make one valve, the RM inventory is shown as 7,000/0.34= 20,588.
- Show under each process step some key data: Number of operators and machines, CT (cycle time), Processing time (if ≠CT), C/O (changeover time), Lot size or Number of changeovers per week, OEE (overall equipment effectiveness), Available time per day, FPY (first-pass yield).
Step 3: Show the material flow type (push, pull, FIFO) using the appropriate icon in between the process steps
Show the information flows linking customer, process steps and suppliers. Each arrow is a separate piece of information that is labeled with its content, transmission format (e.g. EDI, email…) and transmission frequency. Each box is an information-processing agent (it receives some information, processes it, and sends a different information).
Step 4: Convert inventories into waiting time by dividing by average daily demand
- Draw a crenellated timeline under the process, with waiting time on the high parts, and the processing time on the low parts (under each process box).
- Add all the times on the longest branch to get lead time.
Step 5: Add a timeline for the information flow prior to the order being released to production
The Value Stream Efficiency% is: (Σ process times / total lead time)*100.
Step 6: The whole team builds the value stream map on a large piece of paper
- Typical dimensions are 5 feet tall by 10 feet long or more
- Discuss what the biggest problems and opportunities are, using “bursts” icons to show where improvements are needed.
Frequently Asked Questions About VSM
Q: How detailed should the information flow be?
It should explain at a glance how we transform information coming from the customer into information that triggers availability of materials and the actual work. There’s no need to go in detail beyond “Customer”, “C/S & scheduling”, “Procurement”, “Supplier” and perhaps “Supervisor” in terms of information processors. What’s important is to identify delays, excessive complexity, etc. Don’t forget to add the 3 labels to every information arrow: what’s being transmitted? How is it transmitted? How often?
Q: How do I deal with a process step being performed by a subcontractor (e.g., coating)?
You typically won’t have the process data, so all you know is how long it takes for a batch to come back after being shipped. If possible, estimate (or ask) the processing time and add it on the timeline. That will allow you to evaluate future state alternatives.
Q: When counting inventory, is it OK to take data from our ERP/MES?
It’s OK to use them for raw material and finished goods, but we recommend doing a physical count for WIP. You can pick up a lot more on the gemba than on a screen!
Q: Should I count all WIP in front of a process step, or only the parts that belong to the value stream product family?
In a lean transformation, aim for a value stream organization. That means work centers should be dedicated to a value stream, in which case the question won’t come up. But if you do have a shared process, then you should count all WIP in the queue, even for products outside of your value stream (however, see next question).
Q: How do I convert inventory into days at a shared process step?
Ideally, you take all the products in inventory and divide it by total demand per day for those products, as that would be more accurate for material waiting time. A less accurate option is to count only the value stream’s products in inventory and divide this number by the daily demand for those products.
Q: Can I measure material waiting time by other means, such as analysis of routing sheets or trackers?
This can be a more accurate method than the “days on hand” calculation, but the data is typically entered by hand, so the key question is “how confident are you about the accuracy of the data?” If the data is good, this a great method because it gives you a distribution of lead times instead of just averages.
Q: Do I count a Team Lead or Material Handler in the number of operators at the process step?
There isn’t a generally accepted answer to this question, but we recommend counting only operators engaged in value-added activities. Numbers appear in the data box for two reasons: they can be used to draw the timeline, or they’ll be different in the future state vs. the current state. Normally, the number of operators will change, but if you have a Team Lead today, you’ll still have one a year from now. Whatever you choose, what really matters is consistency.
What's Next?
- The word Value Stream Map or VSM is by convention reserved for describing the current state.
- A Value Stream Design or VSD map, is a statement of a future state.
- The next step, having completed a VSM with a shared understanding of the current state among all the stakeholders, is to work together to define an Ideal State VSD.
- This is something a few years in the future, where all the lean principles are applied and ambitious, but reasonable (no science fiction) goals are set.
- Next, based on the VSM and Ideal VSD, the team defines what they think they can achieve 12 months from now. This is called the Target VSD.
- A VSM-Target VSD gap analysis is performed, from which a list of improvement projects is defined.
- For each project, a charter is developed and a C.I. Plan is developed:
- Sequence and schedule for each project
- Individual and cumulative goals are defined from the project charters, covering Quality, Cost, Delivery and People metrics.
- When you perform the VSD, VSD, Gap Analysis and Project Planning in one event, this is called Value Stream Planning (usually over two or three days).
- The goal is to integrate VSPs with strategy deployment, at a consistent cadence.
