The "Return and Rebirth" of the PCBA Production Line: A Lean Practice from Problem to Solution
In the past two weeks, my work rhythm has been like a wound-up clock, all centered around the "relocation and reconstruction" of the PCBA production line. Saying "reconstruction" is not an exaggeration. This production line was moved from the old factory to the new one at the beginning of the year with the intention of "building a professional base," but it ended up causing a whole bunch of problems. Now that it's being moved back, what we need to do is not to "duplicate it as it was," but to turn all the previous pain points into areas for improvement.
From Relocation to Return Relocation: The Unhidden Wrong Relocations
We are a typical example of "one factory with two bases": The old factory is responsible for final assembly. PCBA (Printed Circuit Board Assembly) used to be the core process of the old factory. At the beginning of the year, a sudden decision was made: Move the PCBA production line to the new factory, with the reason being "The new factory has more spacious workshops and can build a professional PCBA base."
But the problem soon emerged:
The plan coordination has failed: Two factories share a single planning center, but the PCBA output of the new factory can never keep up with the assembly rhythm of the old factory. For example, in March, the mobile phone motherboard assembly line of the old factory stopped operating three times due to a shortage of PCBA. Each time the line stopped, overtime work was required to make up for the production capacity. The PCBA from the new factory has to be first put into storage, then loaded onto a vehicle, and then transported to the old factory, which takes at least two hours in between, and the assembly line of the old factory cannot afford to wait.
Efficiency fails to meet the standard: The production line of the new factory is a 20 - meter - long assembly line copied from the old factory. It takes 1.5 hours to change the production line. However, 60% of our orders are small orders of 500 - 1000 pieces. The cost of changing the production line is extremely high. For a single small order, the time spent on changing the production line alone accounts for 30% of the total production time.
ERP bottleneck: The new factory uses the same ERP as the old factory. However, the production line layout is scattered, and material inbound and outbound operations need to be recorded across different factory areas. For example, when materials are transported from the old factory to the new one, they need to be recorded as "outbound" in the old factory first and then as "inbound" in the new factory. Each time, it takes a long time for verification, and the error rate of orders is 20% higher than before.
Costs are soaring: When PCBA is transported from the new factory to the old factory, three trucks make trips every day. The additional monthly cost of freight and handling losses amounts to 30,000 yuan.
"Homework" before reconstruction: Determine the direction with data
After deciding to move back to the old factory, I didn't rush to dismantle the equipment. Instead, I first did three things to "lay the foundation":
1. Ascertain order requirements: I asked the PMC for the order details of the past six months. There are 15 types of products. The minimum quantity per batch is 50 pieces and the maximum is 8,000 pieces. The line change frequency is 5 - 6 times a week. The conclusion is clear: We need a small and flexible production line rather than a large and clumsy long assembly line.
2. Measure on-site data: Before going to the new factory to dismantle the production line, the dimensions of each piece of equipment (the insertion machine is 1.2 meters, the soldering dip machine is 2 meters, and the wave soldering machine is 1.8 meters), the operating radius of the workers (with an average of 0.6 meters), and the location of the existing material area (the original material area is 5 meters away from the production line) were measured.
3. Determine principles: Based on these data, four core principles for the layout are clearly defined - safety (leave a 1.5 - meter - wide fire - fighting passage), short distance (the distance from materials to the work surface ≤ 1 meter), smooth flow (no cross - moving lines), and one - piece flow (no dropping to the ground from component insertion to touch - up soldering).
Five "lean points" of the layout: Each modification solves an old problem
This layout plan doesn't have any "shams" and is entirely about "precision strikes" against long - standing problems.
1. Unilateral operation: Solve the pain point of "taking a detour for feeding"
The original production line was a bilateral operation. Workers had to walk around to the material rack on the opposite side to load materials, which meant they walked an extra 1.2 kilometers every day. Now it has been changed to a unilateral operation - the material rack is placed directly on the left side of the workbench. Workers can reach out to pick up materials such as resistors and capacitors. The material loading time has been reduced by 20%. Workers said, "We don't have to run back and forth, and our waists don't ache anymore."
2. Inclined material rack: Change "bending down to get materials" to "getting materials conveniently"
Previously, the feeding rack was flat, and workers had to bend down or stand on tiptoe to reach the materials at the bottom. Now, the rack is made into a 15-degree inclined plane, and the opening of the material box is just facing the workers' operating surface. The distance for taking and placing materials has been shortened from 0.8 meters to 0.5 meters. It has been calculated that each worker can reduce 15 bending actions per hour, the fatigue degree is reduced by 30%, and the operation efficiency can be improved by 12%.
3. Split into two 10-meter lines: Solve the problem of "difficulty in replacing lines for small orders"
The original production line was a 20 - meter - long large line. When changing the production, the whole line had to be shut down for 1.5 hours. Now it has been split into two 10 - meter small lines. For small - batch orders, one of the lines is used, and for large - batch orders, both lines run simultaneously. For example, when producing a small order of 500 pieces, the line - changing time has been reduced from 1.5 hours to 40 minutes, and it doesn't affect the regular production of the other line. The production capacity utilization rate has increased by 35%.
4. Connection of conveyor belt connectors and rework soldering: Implement "one-piece flow" and eliminate work-in-process
Previously, the PCBA after the plug-in process had to be packed into turnover boxes, transported to the rework soldering line and then unpacked. The in-process inventory often piled up like a "small mountain" (up to 500 pieces at most). Now, a 1.5-meter-long conveyor belt is used to directly connect the outlet of the tin dipping machine to the inlet of the rework soldering line. The PCBA doesn't need to be placed on the ground, and the in-process inventory has been reduced to less than 50 pieces. This not only reduces the handling losses but also enables real-time detection of plug-in problems (for example, the rework soldering workers can immediately report poor solder joints).
5. Repair and modify the unit line: Solve the waste of "idle running of the assembly line"
The original repair welding was carried out on a "fixed-speed" assembly line. When producing in small batches, most of the line was idle, resulting in wasted electricity and labor costs. Now, three small work units have been set up using lean racks (each unit consists of 2 workers, 1 soldering iron, and 1 material box). When producing a small order of 100 pieces, one unit is activated; when producing 500 pieces, two units are activated. This has reduced costs by 45% compared to the previous method. More importantly, there is the balance rate. Previously, the balance rate of the assembly line was only 70% (there were always workers waiting for materials), but now the unit line can achieve a balance rate of 95% because the materials and personnel in each unit can be adjusted at any time.
It hasn't been implemented yet, but the "effects" can already be "seen"
Now the layout plan has passed the reviews of all departments. The workshop supervisor said, "The two small production lines are just right for our orders." The workers said, "The modification of the inclined fabric rack is very practical." The PMC said, "There's no need to worry about the conflict between small and large orders when scheduling production anymore."
Standing in the empty workshop of the old factory, I held the layout diagram and gestured: The component insertion workers load materials from one side. The PCBA slides to the rework soldering unit via the conveyor belt. Workers don't need to wait for materials. When changing the production line, only one line needs to be adjusted... Although the equipment hasn't been moved back yet, the sense of achievement of "turning theory into an implementable plan" is stronger than that of completing any existing project.
The busyness this week is not about "rushing to complete work", but about "turning previous mistakes into correct experiences". Isn't what we call lean production all about "finding problems, solving problems, and then optimizing problems"? Once the production line gets up and running, I'll share the real efficiency data with you all!