In modern manufacturing, as customers demand more customised product variants to meet their specific needs, the number of small, different parts required for these variations has increased significantly. For example, an industrial drill manufacturer might have 10 products, each with 10 variants. Each product variant contains 10 unique parts, in addition to 30 parts that are common to all products and variants. To produce all these variants, the manufacturer needs to manage between 2,000 and 4,000 parts. Additionally, even parts that seem similar, such as seals, may differ slightly, for instance, by being made from different materials depending on the variant.
If you manufacture many product variants, you are likely familiar with the challenges described above. Your warehousing operation may struggle to manage the balance between unique and common parts as it supports production. Take shims as a simple example. They are small yet critical machine components. There are numerous shim shapes and sizes, which can be difficult to distinguish even for a trained eye. The complexity increases further when considering the wide variety of materials used to make shims. Selecting the correct shims is essential in manufacturing product variants, and that is just one example.
Most product parts are stored or warehoused before use in production. This is especially true for parts that cannot be individually labelled, such as special bolts, small unique components, sub-assemblies, and similar items. These parts must be identified and tracked to ensure they are picked correctly for production. When thousands or even tens of thousands of orders are processed daily, errors caused by haste directly result in defective products and delayed deliveries, triggering a costly domino effect. Therefore, part storage and picking processes must actively prevent picking errors to maintain quality and efficiency
Cost of picking errors
Picking errors are potentially catastrophic in manufacturing because they can unleash an avalanche of margin-eating costs. The most immediate cost is often the use of courier services to urgently deliver the correct parts to customers who have received a product but are unable to use it. But that is just the tip of the iceberg. Handling the claim and identifying the root cause of the error burdens the entire operation. If inventory allows, you may have enough parts to send the correct replacements afterwards. You must absorb the costs of rushed delivery and disruption to regular operations, expenses that you could have avoided. In the worst case, insufficient inventory causes further delays and a cascading impact across the supply chain.
If picking errors are frequent, customer relationships suffer, and you risk losing revenue to competitors. Perhaps one customer received the last, but incorrect, part that another customer needed. Now you have two issues to resolve. How can we avoid this painful "drawing empty" scenario? One option is to maintain a high inventory buffer, although this increases costs. Another is to perform perpetual inventory counting, which is a tedious process and does not address the root cause.
When you discover unavailability too late
The worst scenario occurs when a missing part is discovered during production or while preparing an order for shipment. Even a single unavailable part can stall the entire variant, tying up capital in work-in-progress and blocking inventory that could otherwise complete other variants. Missing parts not only cause inconvenience but also significantly delay and disrupt operations. At best, they waste valuable time; at worst, they bring production or shipment to a complete halt.
The risk of tacit knowledge
In many operations, critical knowledge, such as the location of specific parts or how to identify them, is often held in the minds of experienced employees. When those individuals leave, the tacit knowledge they possess also leaves with them. Without a precise and reliable picking system, the onboarding process for new personnel can be slow and inconsistent. Implementing a robust and straightforward picking process is crucial for maintaining efficiency and operational continuity.
Picking the right parts
Investing in warehouse software and automation is essential to prevent picking errors; however, selecting the right solution is not always straightforward.
Most warehouse systems fail to address the root cause of errors, only reducing the likelihood of mistakes. When a person faces too many choices, errors become inevitable. For example, shelves filled with bins of similar-looking items increase the chance of picking the wrong part. Presenting only one bin to the picker dramatically lowers this risk. Limiting picker options is one of the most effective strategies and likely your best investment.
Enterprise Resource Planning (ERP) systems effectively manage stock data but remain disconnected from physical operations. They rely on manual inputs and predefined processes, enabling them to detect discrepancies but not prevent physical errors. Similarly, Warehouse Management Systems (WMS) track SKU locations and inventory levels but depend heavily on manual accuracy. Neither system can physically prevent a person from selecting the wrong item or incorrect quantity.
The ideal picking environment
In an ideal scenario, the system presents only the part that is physically needed. When that happens, picking errors can be eliminated. Systems like pick-to-light use lights or lasers to guide the picker to the correct item. However, they are not foolproof. Items may be misplaced or picked incorrectly despite the guidance. The process becomes more straightforward and unambiguous when you make only the correct item available.
Picking the right amount
It is not just about picking the correct part. Selecting the proper quantity is equally important. One proven method is weighing. When parts are stored, their weights are recorded along with the total number of units in the container. During picking, the removed parts are weighed again, and the system verifies that the weight matches the expected quantity, ensuring accuracy and avoiding shortages.
Other methods, such as machine vision and RFID, are being explored, but they are not yet practical or reliable enough for commercial use.
Key takeaways: ensuring accuracy in variant manufacturing
- Product variant complexity: Modern manufacturers must manage a massive increase in unique parts and subtle variations, even for small components like shims, which can create critical challenges.
- The cost of picking errors: Mispicked parts can lead to production delays, increased operational costs, lost inventory, and customer dissatisfaction.
- Storage and picking must be foolproof: Traditional warehousing struggles to keep up, particularly with unlabeled, similar-looking parts like bolts and seals.
- Manual processes are vulnerable: Even trained staff can make mistakes; reliance on human memory and "silent knowledge" poses a risk during staff transitions.
- Software alone isn't enough: ERP and WMS systems manage data effectively but don't physically prevent picking errors; they rely on accurate human input.
- Automation is the game-changer: Systems that present only the correct part or guide pickers visually (e.g., pick-to-light) drastically reduce errors.
If you're looking for ways to seriously improve picking accuracy in your manufacturing operation, check out the latest in this critical area: https://www.konecranes.com/equipment/agilon
Vesa Hämetvaara
Director, Agilon Business
Konecranes Agilon
email: vesa.hametvaara@konecranes.com
Teemu Oittinen
Director, Agilon Solution Sales
Konecranes Agilon
email: teemu.oittinen@konecranes.com
