Material “arrived” at the station but the ledger still shortages? Close delivery scans back to WMS

There is one step between 'gave' and 'saw'

The morning meeting was at 7:30. The planner pointed at the screen: 'This batch of materials was delivered yesterday.' The line supervisor handed over a video on his phone: an empty pallet next to the workstation, with the label still in the turnover area. After ten minutes of arguing, someone remembered—the driver had unloaded at the workshop door, the receiving scan wasn’t done, and the system still showed 'in delivery' or was marked 'completed' by the warehouse staff. The account shows it exists, but on-site it’s gone. So, another material request is issued, and 'buffer stock' starts piling up at the line. The more buffer stock there is, the worse the inventory checks hurt, and the less reliable the completeness—this is exactly the interest paid after a failed closed loop.

Materials have arrived at the workstation but are still missing in the accounts; the root cause is rarely summed up by the four words 'warehouse staff being lazy.' When inventory tasks get stuck midway, yet completeness checks are passed as 'available,' both scheduling and on-site reporting will simultaneously lie. The more you are mentioned in the WeChat group, the more it shows that the closed loop is not in the system. The longer the shortage report, the more it often indicates that classification wasn't done and the definition of completion wasn't updated.

There is another kind of illusion: the on-time delivery report looks good because the completion node is set at 'warehouse dispatch' or 'loading,' rather than 'station reception.' The metrics look good, but the production line is still stopped. First, change the definition of completion, then discuss assessment; otherwise, assessment will solidify the wrong feedback loop.

First, divide the material shortages into three categories

If there is only one way to report a material shortage, the countermeasures will definitely be chaotic. Classify them before the meeting:

Export the list of 'Delivered but Not Received' to the site for verification; it would be more useful than going through another round of material reminders. Long-term in-transit items should trigger alerts, rather than relying on manually checking chat records. Areas with frequent repeated material requests are almost always due to in-transit items being invisible—if you can't see them, you request again; if requested too many times, they accumulate on the line, triggering a new 'no space to place' issue.

When the three types are mixed in discussions, procurement, warehousing, and production lines will prescribe solutions according to different causes: procurement speeds up, warehousing sends more, and production lines request again — inventory and the site will only become more chaotic.

The closed loop only has four steps; missing one step makes it fake.

Treating a request for materials as a WeChat message is the starting point of failure. The complete process should be:

1. Demand: Initiated by workstation or work order, including materials, quantity, required time, and target workstation.
2. Ingredients: WMS generates picking tasks, scanning containers/pallets bound with materials.
3. Delivery: Loading/unloading scan, status is in transit, shall not be counted as available at the line side.
4. Receive: The workstation or the line-side warehouse must scan to close the in-transit status before it becomes available to the workstation.

Container codes must be tied to the target workstations to reduce the mentality of 'considered complete once delivered to the workshop door.' Even if batch picking is accurate, if the unloading point does not match the workstation, the production line will still report shortages. For materials from a changeover that have not been fully used, the return rules must be clearly written in advance—who initiates the material return task, how long it must be stored back in the warehouse, whether temporary storage is allowed at the line side, and whether such storage counts as a complete set. Otherwise, the line side will become a second chaotic warehouse: records are in the warehouse, goods are on the line, and inventories cause headaches in both places.

Skipping scanning for urgent materials is possible, but it must be approved and the re-scan deadline set. For cases without a deadline, ‘send first, scan later,’ the discrepancies between accounts and inventory will explode in the inventory month after three months. The proportion of exceptions should also be reported in the weekly meeting: if urgent materials become normal, it indicates a problem with the planning window or safety stock strategy, not that the scanning is ‘too strict.’

Who manages the account: MES and WMS should not operate on separate tracks

During the planning phase, the main responsibilities must be fixed: MES initiates material issuance/return and delivery requests and consumes logistics status; WMS manages the main inventory accounts and operational events, and writes back to MES. The line-side area can be configured as a line-side warehouse or a workstation storage location, but the transactions must be closed-loop, and the person responsible for inventory counting must be clearly defined. Whether it is called a 'line-side warehouse' or a 'workstation storage location' is not important; dual-track accounting is fatal.

Kitting and dispatch must consume the same set of logistics standards. APS shows kitting and line-side shortages — most likely because available quantities were taken up by in-transit items that haven't been received yet, or MES/WMS each recorded them separately. If the reported consumption doesn't match the line-side receipt, process material traceability will also break: you can say what materials were used, but you can't match when they were delivered to the workstation.

The supply-side shortage plan should be based on the actual shortage classification (out of stock vs in transit), not meeting screenshots. Otherwise, procurement will rush for materials that 'are actually already on the way,' while what should be replenished in the warehouse won't be replenished.

When pure manual lines and warehouse production are in the same space and visual flow is clear, you can first simplify storage locations before implementing a closed loop; without a PDA, yet insisting on committing to JIT, it will only create more 'system shows it was sent.' First solve the discipline of terminals and scanning, then commit to rhythm-based feeding.

How to connect this chain with JEDP-MES

JEDP-MES internal logistics allocates materials according to the picking list and delivers them based on line-side demand, supporting scanning for loading/unloading and workstation binding; it collaborates with WMS for material call delivery and visualizing shortages. The positioning of the process and in-process logistics module is to make logistics a constraint for plan achievement, rather than temporary on-site material urging. Acceptance sampling should include: material call—delivery—line-side receipt consistent with WMS inventory events; long-term in-transit materials and repeated calls can be observed.

Phase one suggests only opening up one production line: stabilize the closed loop of material requesting and receiving first, then add shortage classification and material return, and finally connect the kit verification to the same status. Rolling it out across the entire workshop without receiving scans is equivalent to copying false records to every line. Phase two should then align with work order interception/exceptions: items not received cannot be considered as released in full, and exceptions must be explainable.

Reporting work and logistics should not speak in isolation: if batch collection for material feeding is enabled, it should be able to respond to online receiving events; if it does not match, first check the closed loop, then check the traceability depth.

Line-side inventory strategy: clarify first, then discuss lean

When the debate over 'whether to build a warehouse along the line' is ongoing, first answer four more specific questions:

1. Availability definition: Available workstations = Received; Should in-transit items be counted towards completeness? By default, they should not be counted.
2. Inventory ResponsibilityWho bears the differences at the production line? The warehouse keeper, the line-side logistics staff, or the team? Locations with no one responsible are definitely dirty.
3. Water Level and ReplenishmentIs the material requisition based on the work order requirements, or on the minimum stock level at the line? If mixed use is allowed, switching rules must be written.
4. Changeover cleanup: How long must unused materials be returned? If not returned after the deadline, will new work order material requests be blocked?

Under the lean slogan of 'deliver a little more to prevent line stoppages,' if the rules for not exceeding water levels and for returns are not followed, it will only move the warehouse to the aisle. The premise of JIT promises is receiving a closed loop and visibility of goods in transit; when the premise is not established, the honest approach is to lengthen the supply window rather than pretending that JIT is already being achieved.

Containers and paths also need to be standardized: whether the same material is allowed to be delivered in both bulk and full-case, whether container codes must be scanned before unloading, and whether unloading points and workstation codes are one-to-one. When the path is unclear, delivery personnel will choose to "place it where everyone can see"—which is exactly where inventory discrepancies are most likely to occur. The connection with wave picking is: accurate picking within the warehouse only solves the ingredient preparation segment; the workshop segment still relies on loading/receiving events to close the loop. Looking at the two segment indicators separately is the only way to avoid the disconnect of "high picking accuracy, yet still material shortages on the production line." Returns to the warehouse must also enter the task pool: without proper return, the line-side inventory levels will always be inaccurate.

Minimum verification this week

Don't buy more boards yet. Do these three things:

1. Export the list of 'Delivered but not received' from the system for the past 7 days, and check on-site item by item where the goods are and at what status they are stuck.
2. Pick a work order that is currently lacking materials, classify it using the three categories method mentioned above—see if the morning meeting's conclusion correctly identified the problem; record it as a 'misdiagnosis' case for the weekly meeting.
3. Confirm with the warehouse manager: whether the 'Delivery Complete' button can be pressed when there is no workstation receiving; if it can, first modify the definition and permissions of 'Complete', then discuss expansion and the JIT slogan.

If the backlog of 'not received' items obviously decreases within a week and duplicate material requests are reduced, then we can discuss the second line. If the backlog does not move, first check the scanning points (entrance vs workstation), container bindings, and main responsibility table. Do not assign more delivery personnel to 'deliver more'—delivering more will only worsen the secondary warehouse at the line side. Being able to redefine 'completion' to be at workstation receipt alone is worth the cost for this week. Also, make sure planners can see in-transit and received items separately in the kit-check view, so that next week's morning meeting won't continue to argue over the same set of false kit completions.

• MES manufacturing solution
• APS Dispatch Reality
• Scan code to report work collection point
• Wave Picking Accuracy