RSC Colour: Extra Dark Blue

What MES can realistically do to reduce missing parts in kitting

An MES can reduce missing parts primarily by enforcing process discipline, improving visibility of material movements, and creating traceability, not by “automatically fixing” kitting. It can require operators to scan parts, record where every kit is staged, and block orders from moving forward if required components are not confirmed. The effectiveness of this depends on barcode or RFID coverage, accurate master data, and stable interfaces with ERP or warehouse systems. In most plants, MES complements—rather than replaces—warehouse controls, physical labeling, and kitting procedures.

In a typical setup, MES receives the kit requirements from ERP or planning systems and presents an operator with a controlled picking sequence. The operator must confirm each component (by scan or manual confirmation) before MES allows the kit to be released. The system logs which operator picked which part, when, and for which order, providing traceability when parts appear missing later. This does not stop someone from physically misplacing a part, but it does narrow down when and where it likely occurred.

Scan-based picking and verification controls

The most direct MES control is scan-based picking: the system requires each bin, location, or part to be scanned before a kit can be closed. This ensures that the parts linked to a kit in the system match what was physically handled, at least to the level of granularity that the labeling and identification scheme supports. When integrated with inventory data, MES can warn the operator if the wrong part number, revision, or batch is picked, or if a location is out of stock.

MES can also enforce dual verification or independent verification steps for high-risk kits, such as safety-critical or configuration-sensitive assemblies. For example, a second operator or supervisor may be required to confirm the kit contents in MES before the kit is released from the kitting area. These checks reduce the probability of missing or wrong parts, but they increase cycle time and must be selectively applied based on risk and throughput constraints.

Location tracking, staging, and status visibility

MES can model kitting areas, staging racks, and point-of-use locations as explicit locations in the system, and require that kits be moved between them using transactions. Each move (e.g., from central kitting to line-side rack) is logged, so supervisors can see where a kit is supposed to be at any time. If a part is reported missing at the line, MES history helps determine whether the part ever left kitting, was moved in a partial kit, or was reallocated.

However, the fidelity of this tracking depends on how granular the locations are modeled and how reliably operators execute the move transactions. Coarse locations like “Kitting Area A” provide limited diagnostic value when parts go missing within that zone. Fine-grained locations (specific rack/shelf/slot) improve traceability but add scanning workload and are often resisted if kitting takt is tight. Plants must balance operational burden against the desired level of control.

Integration with ERP/WMS and inventory accuracy

MES alone cannot prevent missing parts if the underlying inventory data in ERP or WMS is wrong or delayed. If upstream systems show stock that does not exist physically, MES will still issue pick lists that cannot be fulfilled, forcing kitting staff into workarounds that bypass controls. Conversely, if MES is not integrated and relies on manual imports, timing gaps and data mismatches can create confusion about whether a part truly exists or has already been allocated to another kit.

A robust integration pattern typically involves ERP/WMS remaining the system of record for on-hand inventory, while MES manages kit-level allocation and consumption. MES can reserve quantities for specific orders, preventing double-issuing the same stock to multiple kits. When integration is weak or absent, operators often fall back to informal practices (shadow spreadsheets, physical tallies), which undermine the control that MES could otherwise provide.

Traceability, genealogy, and investigations when parts go missing

One of MES’s main contributions is not preventing every incident, but making it easier to investigate and contain issues when they occur. Item-level or lot-level genealogy links each part to its kit, work order, and eventual assembly, so when a missing or suspect part is discovered, the scope of affected work can be identified. The event log shows who handled the kit, which stations it passed through, and which exceptions were overridden.

This traceability is only as good as the labeling scheme and data capture discipline. If several distinct physical parts share the same generic identifier in MES, you may know that “a part” was issued, but not which one was lost or misused. Where missing parts create serious risk, plants often justify the effort of serial-level tracking in MES, despite higher labeling and data volumes. In lower-risk contexts, a coarser lot-level traceability may be more practical but will limit root cause analysis precision.

Error-proofing, alerts, and escalation workflows

MES can support error-proofing by blocking work from progressing if kitted components are incomplete or unverified. For example, downstream operations may not start until MES confirms that all mandatory kit items are present and within specification (including expiry dates or revision levels where relevant). This prevents some classes of rework and line stoppages caused by missing parts discovered too late in the process.

Additionally, MES can be configured to raise alerts when unusual patterns occur—such as frequent kit re-openings, repeated short-picks for the same item, or a spike in adjustments from kitting staff. These signals can be routed to supervisors or material planners to investigate systemic issues rather than treating each missing part as an isolated event. The usefulness of such analytics depends on consistent event logging and may require tuning to avoid alert fatigue.

Physical and procedural controls MES cannot replace

MES cannot replace basic physical and procedural controls, such as secure storage, clear labeling, 5S in kitting areas, and controlled access to high-value or safety-critical parts. If parts can be picked directly from bulk storage without scanning or if operators frequently “borrow” parts between kits without recording it, the MES data will diverge from reality. In such environments, missing parts remain common regardless of the software.

Similarly, MES does not solve problems caused by poor layout, overloaded kitting staff, or frequent last-minute engineering changes that invalidate kits. In regulated environments, changes to BOMs and work instructions must go through formal change control, and MES updates must be synchronized carefully with ERP and documentation systems. If that synchronization lags, kits may be prepared according to obsolete instructions, leading to apparent shortages or wrong parts when orders reach the line.

Brownfield coexistence and incremental deployment

In most brownfield factories, kitting is already managed partly by WMS or ERP and partly by informal practices. Trying to replace all of that with a new MES in one step often fails due to integration complexity, validation overhead, limited downtime, and resistance from experienced operators. A more practical approach is to start by digitizing a subset of kitting operations (for example, critical programs, high-value components, or specific kitting cells) and gradually expanding.

Coexistence typically means that some kits are still prepared using legacy methods while others follow the MES process. This hybrid state can expose inconsistencies and requires clear rules about which system is authoritative for each area or product family. Plants need disciplined change management and training to avoid confusion, especially in regulated contexts where process descriptions and validation evidence must match what is actually happening on the floor.

Specific considerations for regulated and aerospace-grade environments

In aerospace and similar high-regulation sectors, missing or mis-kitted parts have implications beyond cost and schedule—they can undermine traceability and configuration control. Introducing or changing MES functionality around kitting often requires validation, documentation updates, and sometimes customer or authority notification. These burdens make frequent changes unattractive and push plants toward stable, well-understood workflows rather than experimental automation.

Full replacement of existing kitting systems and processes with an MES-led approach is often constrained by legacy equipment, qualified processes, and the risk of extended downtime. Instead, MES is usually layered on top of existing controls: enforcing scan discipline, adding genealogy, and improving visibility without discarding proven warehouse or kitting procedures. This incremental approach may feel conservative but aligns better with long equipment lifecycles, certification obligations, and the need for robust change control.

Unique characteristics of aerospace inventory

Inventory accuracy is harder in aerospace because the items being tracked are high value, safety critical, and subject to strict traceability requirements. Instead of tracking pallets of identical parts, you are often tracking individual serial numbers, heat lots, and configuration states. A single line item in the ERP can represent many units, each with different certifications, storage conditions, or applicability limits. This turns what is a basic quantity problem in other industries into a combined quantity, identity, and pedigree problem in aerospace.

Aerospace bills of material typically include complex alternates, effectivity ranges, and life limits, so “having the right part” is not as simple as matching a part number. Two items with the same part number may not be substitutable due to different revisions, suppliers, or approvals. That makes inventory accuracy not just about count, but about whether each stock-keeping unit can actually be used on a given assembly or aircraft.

Serialization, traceability, and paperwork load

Many aerospace parts are serialized or at least lot-controlled, and those identifiers must remain traceable from supplier through manufacturing, test, and in-service. Each movement of a serialized part should update multiple systems: ERP/MRP, MES, QMS, and sometimes separate serialization or repair tracking tools. When these systems are loosely integrated, every transfer becomes a chance for misalignment between physical inventory and records.

Paper-based processes and disconnected scanners are still common because of validation burden and certification constraints, especially around controlled documents. Travelers, certificates of conformity, and inspection records often move with the parts and are keyed in later, if at all. Delays and manual entry errors in this paperwork create timing gaps where inventory exists physically but not yet digitally, or vice versa. Over time, these small mismatches compound into larger inventory accuracy issues.

Regulatory and quality constraints that limit simplification

In less regulated industries, teams can streamline inventory practices, consolidate SKUs, or relax controls to reduce complexity. Aerospace manufacturers often cannot do this without triggering requalification, regulatory review, or customer approval processes. Labeling, storage, and handling rules are constrained by specifications and contracts, which can prevent seemingly simple fixes like re-binning materials or changing how parts are grouped.

Quality and airworthiness requirements also discourage aggressive cycle-count or rework practices that would otherwise clean up data. For example, scrapping or re-identifying ambiguous inventory may require formal material review boards and extensive documentation. This slows down correction of obvious errors and increases the temptation for local workarounds that bypass formal inventory adjustments.

Engineering change and configuration complexity

Engineering change is a major driver of inventory complexity in aerospace. Design updates, service bulletins, and customer-specific configurations all shift which inventory is usable, where, and under what conditions. Parts that were fully usable last month may become limited to certain configurations or require rework before use. If configuration rules in the ERP/MES do not keep pace with engineering changes, the same physical inventory can be represented very differently in different systems.

Configuration-managed products also mean that a single finished item (an engine, avionics unit, or structure) can exist in multiple approved configurations across a long service life. Subcomponents may be swapped, repaired, or upgraded many times, and each of these events changes the effective inventory and its pedigree. Maintaining accurate inventory across new-build, spare parts, and repair/overhaul flows requires discipline that is harder to sustain than in simpler, once-and-done product lifecycles.

Brownfield system landscapes and integration debt

Most aerospace plants operate with a patchwork of legacy ERP, MES, PLM, and QMS systems that have grown over decades. These systems may encode different units of measure, location hierarchies, or part numbering schemes, making consistent inventory representation difficult. When inventory moves across organizational or system boundaries (e.g., from a repair shop to final assembly), reconciliation often relies on spreadsheets, email, or manual uploads.

Replacing these systems wholesale is rarely practical due to validation cost, change-control overhead, and downtime risk. As a result, inventory accuracy improvements must coexist with existing tools and interfaces, which can limit how far you can automate or centralize. Point-to-point integrations and tactical fixes accumulate over time, introducing subtle mismatches in how quantity, status, and location are tracked. These structural constraints mean that even well-designed process improvements may not fully eliminate discrepancies.

Complex storage rules, shelf life, and special handling

Aerospace materials often include chemicals, composites, and life-limited components that require specific storage conditions and strict shelf life control. Inventory records must capture not just how much you have, but remaining life, exposure history, and storage conditions. When these attributes are tracked in separate systems or on local logs, it becomes easy for the main inventory record to fall out of sync with reality.

Kitting and staging add another layer of complexity. Parts are frequently pulled from bulk storage into kits for specific orders or aircraft tails, then partially returned, scrapped, or reassigned. If the kitting and de-kitting processes are not tightly controlled and systemized, inventory tends to fragment into locations and statuses that are opaque to the main ERP. This is fundamentally different from simpler, one-way flows commonly seen in high-volume consumer manufacturing.

Human factors and local workarounds

Because of schedule pressure and the cost of line stoppages, operators and planners in aerospace will often solve local problems first and update systems later, if at all. This might mean borrowing parts across work orders, reassigning serials, or holding material in informal buffer locations. These behaviors are understandable in context but directly undermine formal inventory accuracy.

The training burden is also higher: staff must understand not only how to move parts but also the implications for serial tracking, configuration, and certification. When processes are complex, system usability is poor, and feedback cycles are slow, people rationally prioritize getting hardware built over perfect record-keeping. Over years, these small, rational decisions create systemic inventory issues that are much harder to unwind than simple counting mistakes.

What this means for improving inventory accuracy in aerospace

Improving inventory accuracy in aerospace typically requires addressing both data and process, within the constraints of existing validated systems. Efforts that work well in other industries—such as rapid system replacement, SKU simplification, or aggressive re-labeling—often run into qualification, certification, and downtime barriers. Instead, gains tend to come from targeted integration, better event capture at the point of work, and incremental tightening of kitting, returns, and scrap processes.

Expect diminishing returns: moving from poor to acceptable accuracy is feasible with disciplined basics, but achieving near-perfect accuracy is difficult when serialization, configuration, and long lifecycles are involved. Any initiative should explicitly account for brownfield reality, multi-system alignment, and the need to adjust human behaviors that have grown around the current constraints. Without that, projects risk becoming one-time inventory cleanups rather than sustained improvements.