Bottleneck Improvement Case Study: From "Line Waiting for Materials" to "Capacity Leap" in an Electronics Manufacturing Company

Abstract: Orders increased by 30%, but production capacity remained stagnant—longer production lines became more congested, and more overtime only led to more chaos. This was not due to a lack of equipment, but rather a bottleneck constraining the system. This article thoroughly reviews a bottleneck improvement project in an electronics manufacturing company: from using data to identify the true constraint process, to increasing bottleneck output by about 40% without adding any new equipment, and then to the second round of improvements after the bottleneck shifted, all connected by the five-step method of the Theory of Constraints (TOC) to create a replicable path for capacity enhancement.

1. Order Growth Hits Capacity Ceiling: On-Time Delivery Rate Falls Below 70%

A certain electronics manufacturing company (hereinafter referred to as Company A) specializes in the production of industrial controllers, including inverters, servo drives, and power modules, with its main customers being equipment manufacturers. In the first half of 2025, Company A secured two major customer orders, increasing its monthly order volume by about 35% compared to the same period last year. This should have been a good thing, but the production department was not smiling—order growth brought not profit, but a series of delivery crises.

The problem first surfaced in the delivery end. Previously, with a monthly shipment of 1200 units, the on-time delivery rate consistently stayed above 90%. After the order volume increased to 1600 units, the on-time delivery rate steadily declined, falling below 70% by June 2025. Customer complaints were constant, and sales had to frequently expedite orders, which disrupted the production rhythm, creating a vicious cycle of "the more expedited, the more chaotic, and the slower the production."

The management's first reaction was to "add people, add overtime, and add equipment": two additional operators were added to the testing process, two hours of overtime were scheduled daily for the final assembly line, and even two new testing devices were requested for procurement—although the budget was approved, the equipment would not arrive for three months, which was too far off to address the immediate need.

More puzzling was that after adding people and overtime, there was almost no change in capacity. The Overall Equipment Effectiveness (OEE) reports showed that all processes were operating at around 70%, suggesting "room for improvement," but the overall output did not increase. The issue for Company A was that everyone was focused on the efficiency of their own processes, and no one was addressing the fundamental question—where is the true bottleneck in the entire system?

2. Bottleneck Identification: The Blockage Is Not in the Crowded Final Assembly Line, but in the "Unnoticed" Testing Area

The improvement team did not continue to "sprinkle pepper" but instead introduced the first step of the Theory of Constraints (TOC): identifying the constraint. They spent two weeks doing three things.

First, they drew a value stream map (VSM), marking the cycle time, work-in-progress (WIP) quantity, and first-pass yield for each process from material entry to finished goods storage. The results were clear: the final assembly line had a cycle time of about 4 minutes per unit, the packaging process 3.5 minutes per unit, while the whole machine testing process had a cycle time of 10 minutes per unit—over 600 units of WIP were piling up in front of the testing area, occupying nearly half of the workshop's temporary storage area.

Second, they calculated the capacity load. Based on the current month's order requirements, the available capacity of the testing process was only 78% of the demand, making it the only process in the factory with a "capacity gap." Other processes, even with the most conservative estimates, had surplus capacities of over 20%. Although the final assembly line had more people and looked busy, its output ceiling was firmly held down by the testing area—no matter how much the final assembly line produced, if the testing area could not handle it, the WIP would only pile up higher.

Third, they observed the bottleneck process on-site. The improvement team followed the testing area for three consecutive days and discovered a counterintuitive phenomenon: the equipment utilization rate in the bottleneck process was only 65%. Theoretically, it should have been running at full capacity, but in practice, it spent a lot of time "waiting"—waiting for materials from the previous process, waiting for tooling turnover, waiting for changeovers, and waiting for meal breaks. The "idle" bottleneck was precisely the "blockage" of the entire system.

This discovery overturned the management's intuition that "buying more equipment would solve the problem": the bottleneck process was not even using its full existing capacity, and buying more equipment would only result in more idle machines.

3. Maximizing the Bottleneck: Increasing Bottleneck Output by About 40% Without Adding Any Equipment

The second step of the Theory of Constraints is to "maximize the constraint"—improve the bottleneck process without additional investment. The improvement team implemented five measures aimed at the goal of "making every minute of the bottleneck productive."

First, set a "quality gate" before the bottleneck. Previously, the testing area often detected nonconforming products and sent them back to the final assembly line for rework, causing rework items to re-enter the queue and waste bottleneck capacity. After the improvement, a 100% simple functional pre-inspection was added at the end of the final assembly line to intercept obvious nonconformities before they entered the testing area. This alone reduced the number of rework items received by the testing area by about 60%.

Second, reduce ineffective time within the bottleneck. The testing program had a 40-second waiting redundancy left over from early compatibility with old products. After optimizing the testing program, the single-unit testing time was reduced from 10 minutes to 8.5 minutes. Additionally, the testing area was changed to "staggered meal breaks," with no downtime during lunch and operators taking turns to eat—this added nearly an hour of productive time each day.

Third, quick changeovers. The testing area had to switch between multiple models daily, and each changeover previously required a 35-minute shutdown. The improvement team standardized the changeover actions into a checklist and moved some adjustment work to the last few units of the previous batch, reducing the changeover time to within 15 minutes, thereby saving about 40 minutes of downtime each day.

Fourth, preventive maintenance of bottleneck equipment. Previously, testing equipment was only repaired when it broke down, causing a shutdown of two to three hours per incident. After implementing a 10-minute pre-shift inspection and a weekly preventive maintenance schedule, the testing area's downtime due to equipment failure decreased by 70% that month.

Fifth, move inspection from after the bottleneck to before the bottleneck. Previously, the final product sampling inspection was scheduled after the testing area, and when batch issues were discovered, the entire testing area had already wasted two hours. By moving the sampling inspection to the end of the final assembly line and combining it with the pre-inspection, problems were identified before they entered the bottleneck, avoiding the waste of bottleneck capacity.

These five measures had immediate effects. One month later, the testing area's daily output increased from 120 units to 168 units, a 30% to 40% improvement—without adding any new equipment or permanent staff, just by addressing the "leaks" and "waits" in the bottleneck process. The WIP inventory in the workshop decreased from over 600 units to less than 200 units, and the on-time delivery rate rebounded to over 95% by the third month.

Notably, while implementing these measures, Company A also changed a seemingly minor performance metric: the performance of the testing area was changed from "equipment utilization rate" to "number of bottleneck units produced." Previously, operators aimed for "machines always running," even if they were processing rework items or trial production items, which made the utilization rate look good but did not increase effective output. After changing to output-based performance metrics, everyone began to actively reject ineffective expedited orders and prioritize the testing of order products, aligning individual behavior with the system's goals. This detail reminds us that bottleneck improvement is not just a technical issue but also a management mechanism issue—where the metrics point, the team's attention follows.

4. Bottleneck Shift: Improvement Is Not a One-Time Deal, but a Continuous Relay Race

After the capacity increase, the improvement team did not celebrate for long—they soon discovered that the testing area was no longer the bottleneck, and the new constraint appeared in the packaging and shipping area.

The packaging area had only one packaging line, with a daily handling capacity of 140 units, which had previously been "protected" by the testing area and seemed more than sufficient. After the testing area's capacity increased to 168 units, the packaging area immediately became the new bottleneck, with finished products queuing up in front of it. This is precisely the cycle of the third to fifth steps of the Theory of Constraints: accommodating the bottleneck, improving the bottleneck, and then returning to the first step to re-identify the constraint.

The improvement team replicated the same methods in the packaging area: adding a simple packaging station, standardizing packaging operations, and moving some packaging actions to the waiting period after testing completion... After three weeks, the packaging area's daily handling capacity increased to 180 units, and the system bottleneck shifted to the outbound logistics process. Each round of improvement followed the same logic: first, identify the most constraining process, then focus efforts on breaking it, and move on to the next.

This cycle brought comprehensive benefits. By the fourth quarter of 2025, Company A's monthly shipment volume stabilized at over 1650 units, a 37% increase from before the improvement. Overtime hours actually decreased because the system's output no longer relied on "everyone working overtime" to maintain. The WIP inventory's capital occupation decreased by about 45%, with the semi-finished products previously piled up in front of the testing area turning into real cash flow. The average delivery cycle shortened from 21 days to 13 days, and customer satisfaction significantly improved. More importantly, Company A established a routine mechanism of "finding bottlenecks—maximizing bottlenecks—upgrading bottlenecks," with the production planning department leading a monthly bottleneck review, transforming passive firefighting into proactive improvement.

5. A Replicable Five-Step Method for Bottleneck Improvement

From Company A's case, a five-step method can be distilled, which any manufacturing company can directly apply.

Step 1, Identify the Constraint. Do not rely on intuition to judge "which process is the busiest," but use data: draw a value stream map, calculate the capacity load rate of each process, and see where the WIP is piling up the most. The busiest process is often not the bottleneck—the true constraint is the process with the "longest cycle time and highest load," even if it appears unremarkable.

Step 2, Maximize the Constraint. Focus on the bottleneck: set a quality gate before the bottleneck, reduce downtime and waiting time in the bottleneck, compress bottleneck operation time, move inspections forward, and stagger shifts to keep the bottleneck running. The principle is simple—every minute of the bottleneck must produce effective products, fully utilizing existing capacity before considering investment.

Step 3, Subordinate to the Constraint. Align the cycle times of non-bottleneck processes with the bottleneck, avoiding the pursuit of local efficiency. The final assembly line does not need to overproduce, as overproduction only increases WIP. Using surplus capacity to support the bottleneck, maintain equipment, and train multi-skilled workers is a more beneficial approach for the system as a whole.

Step 4, Elevate the Constraint. Consider increasing investment—adding equipment, personnel, and capacity—only when the bottleneck has been maximized and still cannot meet demand. Note the sequence—first maximize, then elevate; otherwise, the capacity bought with money will be wasted by "waits" and "leaks."

Step 5, Return to Step 1. After the bottleneck is elevated, the constraint will shift to another process, requiring re-identification and re-cycling. Improvement is a relay race, not a one-time deal; companies need to establish a regular bottleneck review mechanism, making the Theory of Constraints a routine part of daily operations.

6. Final Thoughts

Looking back at Company A's improvement journey, the most valuable lesson is not the five specific measures, but the shift in perspective: from "every process striving" to "first, let the system bottleneck strive." Full-scale overtime and comprehensive improvements may seem like a lot of effort, but the system's output is determined by the bottleneck—without breaking the bottleneck, other processes' efforts only result in higher WIP.

The Theory of Constraints reminds us that the output ceiling of any system is determined by its weakest link. Find it, maximize it, elevate it, and then move on to the next—this cycle itself is the simplest form of continuous improvement. Improvement does not have to start with a "major transformation"; starting with the utilization rate of a bottleneck process is enough.

The key to capacity lies not in the busiest process but in the most constraining process—find it, and you find the switch for the entire system's output.

Knowledge code: 7.1.2

Version: v20260809

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