Inventory on the Value Stream Does Not Lie — A Five-Step Diagnostic Method to Identify Process Issues Using Inventory

The general manager of a mechanical processing plant pointed to the warehouse and said, "More than half of the items here should not exist." The quality department, following his instructions, conducted a special inventory cleanup. However, three months later, the inventory levels returned to their original state. Later, a lean consultant was invited. The consultant did not look at reports or ask about policies; instead, they did one thing: they marked all inventory triangles on the value stream map (VSM), arranged them by days and location, and immediately listed the issues. Inventory does not speak, but it does not lie.

1. Why Inventory is the Most Honest Dashboard on the Value Stream

Three things in the production area are most likely to "lie": reports can be filled out beautifully, meetings can be conducted smoothly, and even the andon lights on the equipment can be bypassed artificially. Only inventory does not lie—it piles up there, box by box, tangible and real.

The essence of inventory is a "accumulator" of process issues. If the changeover time is too long, a batch of inventory will accumulate before the changeover point; if the cycle times of upstream and downstream processes are mismatched, the faster process will inevitably create a backlog; if equipment frequently breaks down, there will be a traffic jam after the faulty point; if quality is unstable or there are batch defects, the work-in-progress (WIP) next to the rework area will only increase; if demand forecasting is inaccurate or the delivery cycle is too long, the finished goods warehouse will continue to pile up.

Therefore, in lean manufacturing, there is a saying: inventory is the "speed bump" of the process and also the "warehouse" of issues. Conversely, understanding inventory means understanding the process's ailments. More importantly, inventory is measurable—days, amount, location, and shape can all be quantified, making it more reliable than any verbal description of issues.

2. Five-Step Method: Using Inventory to Identify Process Issues

Step One: Mark All Inventory Points and Calculate the "Inventory Days" for Each

First, return to the value stream map and mark all inventory points, including raw material warehouses, WIP between processes, and finished goods warehouses, with inventory triangles. Do not miss any. Then, for each inventory point, calculate two figures: inventory days (inventory quantity ÷ daily average usage) and inventory amount (quantity × material unit price).

The key in this step is "counting on-site" rather than "extracting from the system." System records often do not match the physical inventory, so on-site inventory counts should be the standard. After a physical inventory count, one company discovered that the system showed 12 days of raw materials, but the actual on-site inventory was 23 days—the discrepancy between the system and the physical inventory itself was the first issue.

Step Two: Determine the Type of Issue Based on the "Card" Position of Inventory

Where inventory stops, the issue lies: if the raw material warehouse is too high, it is often due to unstable supplier delivery, long incoming quality control (IQC) cycles, or large procurement batches; if there is a backlog before a certain process, it indicates that the upstream process is faster than the downstream, or the downstream process frequently breaks down, changes over, or lacks materials; if a large amount of WIP piles up after a bottleneck process, it suggests that the bottleneck capacity is insufficient, and the upstream processes are still producing at the original pace; if the finished goods warehouse is too high, it points to issues with delivery cycles, demand fluctuations, and order forecasting.

Once the location is determined, the major direction of the issue is clear, and there is no need to guess.

Step Three: Read the "Water Level Difference" to Identify the True Bottleneck

Draw the days of inventory at each point as a bar chart, and you can immediately see where the "water level" is the highest. The process with the thickest WIP accumulation is often the bottleneck of the entire value stream—producing faster before the bottleneck is meaningless, as products will only get stuck in front of the bottleneck.

There is another often overlooked signal: if there is almost no inventory after the bottleneck process, it indicates that the downstream is "waiting for materials," and the output of the bottleneck directly determines the entire line's delivery. At this point, improvement resources should be directed entirely to the bottleneck, rather than optimizing the non-bottleneck processes that "look busy."

Step Four: Examine the "Shape" of Inventory to Differentiate the Nature of Waste

The same pile of inventory can have completely different natures. Turnover inventory is formed by normal batch production, and the issue lies in the batch strategy; the corresponding improvement is to reduce batch sizes and shorten changeover times. Safety inventory is set to cope with fluctuations, and the issue lies in the source of fluctuations; the corresponding improvement is to stabilize equipment, quality, and delivery. Stagnant inventory is inventory that has not moved for more than three months, and the issue lies in demand disappearance, design changes, or quality stagnation; the corresponding improvement is to review and dispose of it, and control new additions.

Many companies manage the three types of inventory together, resulting in higher and higher safety inventory, more and more stagnant inventory, and neglected turnover inventory. Categorize first, then address the issues accordingly.

Step Five: Translate Each Inventory Point into an Improvement Topic List

The final step is to convert the diagnostic results into action. Write a "medical record" for each inventory point: location, days, amount, cause, and improvement direction. For example, "WIP before Process C: 8 days → 45-minute changeover causing a batch of 500 pieces → SMED quick changeover, target 15 minutes, batch size reduced to 200 pieces."

Sort the list by amount and address the points with the highest amount and the clearest cause first. After completing this step, improvements are no longer based on "a feeling that something is wrong," but on "data on the books, positions on the map, and numbers on the line."

3. A Complete Practical Case

A certain automotive parts factory has three production lines, with an annual output value of 600 million yuan, and an inventory turnover cycle of 58 days. The owner has been calling for inventory reduction every year, and the departments blame each other: procurement says sales forecasting is inaccurate, sales says production delivery is too slow, and production says procurement incoming inspection is unstable.

The quality department used the five-step method to create an "inventory water level chart": raw materials 21 days, WIP between machining and heat treatment 9 days, semi-finished products after heat treatment 14 days, finished goods warehouse 11 days, and stagnant inventory of about 3 days. The highest water level point was before the heat treatment process—9 days of WIP, while other processes had only 1 to 2 days.

Following the water level difference, the truth emerged: each changeover of the heat treatment furnace takes 6 hours, and to spread the changeover cost, the upstream process feeds in large batches for 7 days; the heat treatment furnace has a high failure rate, and once it stops, it can be down for two days, causing the WIP to pile up. The root cause of the inventory issue is slow changeover and unstable equipment.

The countermeasures involved three actions: reducing the heat treatment changeover time from 6 hours to 1.5 hours (using SMED and standardizing setup procedures); scheduling preventive maintenance for the equipment in the production plan; and changing the upstream process from "large batch feeding based on forecasts" to "pull production based on actual heat treatment consumption." Six months later, the WIP between processes dropped from 9 days to 2.5 days, the total inventory turnover cycle decreased from 58 days to 37 days, and approximately 35 million yuan in funds were released. Unexpectedly, the defect rate also decreased—less WIP means faster issue exposure, and batch defects are no longer hidden until the end of the month.

4. The Three Most Common Pitfalls When Implementing

Pitfall One: Focusing Only on Amount, Not on Days. High-amount inventory points are not necessarily the most severe issues—high-value materials can occupy a lot of money with just a small inventory, but they may turn over quickly. Days reflect "flow health," while amount reflects "funds tied up." Both should be considered; missing either can lead to misdiagnosis.

Pitfall Two: Reducing Inventory Without Addressing the Root Cause. Inventory is a symptom, not the root cause. Reducing batch sizes without shortening changeover times results in frequent changeovers and reduced capacity; reducing WIP without fixing the equipment leads to material shortages and production stops. Address the root cause, and inventory will naturally decrease; only reducing the numbers will cause inventory to rebound—this is why many companies experience "reduction followed by increase."

Pitfall Three: Cutting Safety Inventory Across the Board. Safety inventory is not the enemy. Blindly cutting safety inventory can lead to production stops due to a single equipment failure or incoming material delay, causing losses far greater than the saved inventory cost. The correct approach is to reduce turnover inventory and stagnant inventory, and to "calibrate" safety inventory—recalculate it using actual fluctuation data, rather than arbitrarily reducing it.

5. One Sentence Summary

Inventory is the accumulator of process issues. Understanding the location, water level, and shape of inventory provides a roadmap for improvement—diagnose first, then act, and inventory will willingly decrease.

Inventory on the value stream does not lie—using inventory to identify process issues, diagnose first, then act, to prevent improvement from rebounding.

Knowledge code: 7.1.1

Version: v20260826

Author: Quality Think Tank Quality Think Tank is dedicated to providing systematic professional knowledge, methodologies, and practical tools for quality management practitioners, helping companies continuously improve their quality capabilities.