Eliminating “Knowledge Silos” in Thai Factories: Transforming Veteran Intuition into Data and Standardized Work

Target Readers: Factory managers, site managers, and manufacturing department managers at Japanese manufacturers operating in Thailand and ASEAN, as well as executives and administrative staff at local subsidiaries. This article is especially relevant to those who feel the following challenges: “When a veteran retires or is transferred back to Japan, the floor falls apart”; “We can’t explain in numbers why good products and defective products are being produced”; or “Work standards exist, but no one follows them.”

Anyone who has been involved with factories in Thailand for a long time has probably experienced a scene like this at least once. When a machine starts acting slightly off, a veteran technician listens to the sound and declares, “It’s about time for maintenance.” Or when a product’s finish varies subtly from lot to lot, the veteran explains it in a single sentence: “That lot was made on a hot day.” The factory runs because of that veteran’s intuition and experience — this is true not only in Thailand but at factories around the world.

However, as of 2026, many Japanese manufacturers are beginning to feel the full weight of how serious a risk that “knowledge concentration in individuals” truly is. Retirement, repatriation, job changes, health issues — the reasons vary, but the moment a single veteran leaves, quality becomes unstable, changeover times stretch, and new employees never seem to develop. In Thai factories, this is further compounded by the language barrier between Japanese and Thai and the high mobility of the workforce, making the problem even more complex.

This article examines the structural causes behind knowledge silos, and explains concrete approaches for converting veteran intuition into data and standardized work. Drawing on TOMAS TECH’s on-the-ground expertise, we will cover how to combine digital tools — IoT, production monitoring, paperless systems, and inventory management — and how to think about implementation decisions with a target of recovering the investment within three years.

1. What Are “Knowledge Silos”? — Understanding the Structural Challenges Unique to Thai Factories

“Knowledge silos” refers to a condition in which specific tasks or skills depend on a limited number of individuals, and operations stall when those individuals are absent. In manufacturing, knowledge silos can be divided into three main layers.

(1) Equipment and Machine Silos

A situation where only “Employee A” can properly handle a particular machine. The machine’s quirks, optimal settings, and troubleshooting procedures accumulate as tacit oral knowledge held by that specific person. Even when equipment manuals exist, “just ask A — it’s faster” becomes the norm on the shop floor, and the manuals become hollow formalities.

(2) Quality Judgment Silos

A situation where veterans make “pass/fail” decisions based on visual inspection and feel. Inspection standards documents may exist, but actual judgment depends on the veteran’s experience. As a result, defect escapes increase on days when the veteran is absent, or conversely, over-inspection worsens yield.

(3) Changeover and Scheduling Silos

A situation where production plan changes in response to order fluctuations, material procurement timing, and line-to-line coordination are all managed inside a particular person’s head. In Thai factories, Japanese managers often carry this role, and when they repatriate or rotate out, handovers frequently fail to function properly.

A key Thailand-specific concern is the high mobility of the workforce. The turnover rate among skilled manufacturing workers in Thailand tends to be higher than in Japan, creating the risk that years of accumulated technical knowledge can be “reset” in a matter of years. In addition, the repatriation cycle of Japanese engineers (typically every three to five years) means that tacit knowledge accumulated in Japanese often disappears without being transferred to Thai staff.

2. The “Hidden Costs” of Knowledge Silos

The costs of knowledge silos rarely show up on financial statements, but they occur every day on the shop floor. Do any of the following situations sound familiar?

The defect rate rises only on days when veterans are absent
It takes six months to a year for a new operator to work independently
Root-cause analysis of quality issues takes too long, and countermeasures end with “individual caution”
Preventive maintenance cannot be planned systematically, and unplanned downtime is frequent
Daily reports and work records vary by individual, making aggregation and analysis unusable
Production results and inventory figures do not match between the system and physical reality

Each of these may look like a small problem in isolation, but they accumulate into significant costs. Rework and scrap costs for defective products, quality claim response costs, increased training and recruitment costs, and rising management costs from increased oversight by the Japanese headquarters — all of these stem from the structural problem of knowledge silos.

Even more serious are the “invisible costs.” When veterans know that operations cannot run without them, their bargaining power increases, and demands for better compensation, interference with work methods, and resistance to change become more likely. The organization becomes locked into dependence on veterans, creating a vicious cycle in which standardization and improvement activities struggle to advance.

3. Turning “Intuition” into Data — The Basic Design of Shop Floor Digitalization

The first step in resolving knowledge silos is to decode the nature of “intuition.” When you carefully break down what veterans are “sensing” when they make decisions, you find that in most cases it is a combination of multiple data points.

For example, the intuition that a machine “sounds wrong” is actually a holistic assessment — built up over years of experience — of subtle changes in vibration, temperature, current draw, and processing time. The intuition that “this lot isn’t going to turn out well” is recognition of a pattern from a combination of factors: material lot, ambient temperature, humidity, and the condition of the previous process.

The approach to converting these into data can be thought of in three broad stages.

Stage 1: Visualization of the Current State

Record the equipment status in real time: running, stopped, in changeover, or in fault. IoT sensors can automatically capture current, vibration, and temperature, but starting with a simple production monitoring system where operators enter the status manually is also effective. The key is to “create a state where anyone looking at the data sees the same thing.”

Stage 2: Standardizing Records

Digitize daily reports, inspection records, and quality records, and unify the input format. With paper daily reports, the way they are written varies by individual, making later aggregation and analysis impossible. Introducing electronic forms that can be entered on tablets or smartphones makes data instantly aggregatable.

Stage 3: Discovering Patterns and Converting Them into Standard Work

Analyze the accumulated data to identify patterns: “Under which conditions are defects most likely?” and “What state indicates that equipment maintenance is needed?” By documenting those patterns as work standards, inspection criteria, and alert rules, the veteran’s “intuition” is transformed into “rules anyone can use.”

4. Using a Production Monitoring System to Make “Equipment Status” Shared Organizational Knowledge

In resolving knowledge silos, one of the most impactful areas is equipment production monitoring. By creating an environment where the current status of every piece of shop floor equipment — running, stopped, in changeover, or in breakdown — can be shared in real time, the “equipment status” that only veterans knew becomes information accessible to the entire organization.

Introducing a production monitoring system brings about the following changes.

Recording and analyzing downtime causes: By logging why equipment stopped every time it does, patterns of recurring problems become visible
Tracking OEE (Overall Equipment Effectiveness): A KPI combining availability, performance rate, and quality rate clarifies improvement priorities
Transitioning to preventive maintenance: Historical downtime data enables planned scheduling of equipment maintenance
Improving handover quality: At shift changes, “what happened during the previous shift” can be verified from data, reducing dependence on verbal handovers

In Thai manufacturing facilities, it is common to see downtime causes recorded by hand on paper daily reports. However, those records depend on each person’s writing style — “motor malfunction,” “electrical issue” — and the inconsistency renders the data useless for later analysis. Transitioning to dropdown-style input in a production monitoring system resolves this problem all at once.

5. Achieving “Standardized Records” Through Paperless Systems (i-Reporter)

Digitizing daily reports, inspection records, and quality checklists is a central element of eliminating knowledge silos. Paper-based forms have three fundamental problems: first, the way they are filled out varies by person; second, they cannot be aggregated in real time; and third, data cannot be accessed without physically locating the paper.

Using a paperless tool like i-Reporter brings about the following changes.

Unified input forms ensure “the same format regardless of who records it”
Dropdown and numeric input fields “eliminate ambiguous expressions”
Records are reflected on the server instantly, making “the latest data visible from anywhere”
Past data can be searched and aggregated easily

Particularly effective for Thai shop floors is the ability to input and view records in both Thai and Japanese. Creating an environment where Japanese managers review information in Japanese and Thai staff enter data in Thai prevents the information gaps caused by the language barrier.

Photo attachment functionality is also important. By attaching photos to quality anomaly records, “what the problem was” can be preserved visually, not just as text. When the next issue arises, the ability to confirm “was there a similar case before?” using images greatly accelerates problem-solving on the shop floor.

Paperless Form Conversion Checklist

Item	Current State (Paper)	State After Digitalization
Uniform daily report format	Varies by individual	Unified form ensures consistent format for everyone
Real-time quality record review	Collect paper, aggregate later	Real-time aggregation and alerts
Accumulation of issue records	Filed in cabinets, difficult to search	Instant keyword search and photo reference
Multilingual support	Japanese only; Thai staff cannot read it	Input and view in Japanese and Thai
Reporting to headquarters	Transfer to Excel for monthly reporting	Auto-aggregation generates reports

6. Putting Numbers to “Inventory Intuition” with an Inventory Management System (PEGASUS)

One aspect of knowledge silos in manufacturing facilities that is often overlooked is inventory management silos. “Leave material procurement to Employee B — she’ll handle it.” “Employee C knows what’s in the warehouse.” When this state persists for a long time, material shortages, excess inventory, and missed billing occur whenever Employees B or C are absent.

Introducing an inventory management system resolves the following knowledge silos.

Transparency of inbound and outbound movements: Records of who moved what, when, and how much remain in the system, removing dependence on the “knowledge in a particular person’s head”
Reorder point alerts: An automatic notification mechanism when inventory falls below a certain level eliminates the need for “Employee B’s intuition” to drive procurement
Standardized lot management: Linking material lots to product lots in the system ensures traceability in the event of a quality issue
Efficient physical inventory: Making discrepancies between physical inventory and system inventory visible allows losses, misappropriation, and input errors to be detected early

In Thai manufacturing facilities, material inventory is often managed in Excel, with multiple staff members each maintaining their own Excel files, leading to a state where “no one knows which Excel is correct.” Migrating to an inventory management system is also an escape from this “distributed Excel chaos.”

In Thai manufacturing, companies exporting products may be required to maintain accurate lot management to qualify for BOI preferential treatment for raw materials. Manual Excel management tends to leave an incomplete audit trail, but implementing an inventory management system enables the accurate output of data required for BOI applications and reports.

7. Documenting and Embedding Standard Work — Eliminating “Only Those Who Know, Know”

Simply collecting data is not enough to resolve knowledge silos. It is essential to build Standard Operating Procedures (SOPs) based on accumulated data and embed them on the shop floor. However, this is where many factories fall into a common trap.

“We created SOPs” — but “nobody uses them.” The root cause of this problem lies in how SOPs are created and communicated. It is common to see paper SOPs sleeping on a shelf, or video SOPs that no one knows the location of.

To make SOPs “live” on the shop floor, the following points are critical.

Visibility of SOPs

SOPs should be posted near the equipment or workstation, or made accessible on a tablet at any point during the work. Electronic form systems like i-Reporter can display the SOP as an on-screen procedure guide during the task.

A System That “Keeps SOPs Updated”

An SOP is not finished when it is first created. Assign clear ownership and procedures for updating SOPs whenever improvements or changes occur. Paper SOPs require printing, distribution, and retrieval of the old version every time they are updated, but digitized SOPs can be viewed in their latest version on all terminals instantly.

Cultivating a “Culture of Using SOPs”

When veterans say “I don’t need to look at the SOP to do this,” it creates resistance among younger workers to referencing SOPs. Managers must consistently demonstrate that “using SOPs is the right way to work.” Combining this with QC circle activities and Kaizen proposal processes creates a mechanism through which the shop floor improves SOPs on its own initiative.

8. Structurally Resolving the “Handover Problem”

In Japanese manufacturing companies in Thailand, the most acute manifestation of knowledge silos is personnel changeover. Japanese manager repatriations, retirements of veteran Thai staff, personnel reassignments to new production lines — many readers have likely experienced a sharp temporary drop in shop floor productivity with each of these changes.

Structurally resolving the handover problem requires creating a state where “knowledge accumulates in the system, not in the person.” The following approaches are effective.

Building a Database “You Can Look Things Up in When You’re Stuck”

Past trouble cases, remediation methods, equipment quirks, and material characteristics — accumulate these electronically and make them searchable by anyone. There is no need to try to build a perfect knowledge base from the start. If day-to-day business records — daily reports, quality records, equipment inspection records — are digitized, they themselves become the knowledge base.

Setting “Shortening the Onboarding Period” as a Numerical Target

Set the time for a new operator to reach independent performance (the onboarding period) as a success metric for eliminating knowledge silos. In an environment where standard work is established and records are accumulated in electronic forms, the quality of OJT (on-the-job training) improves and the onboarding period shortens. A change from “it used to take six months, now it takes three” is directly calculable as a cost reduction.

Redefining the Role of Japanese Managers

The root of the handover problem is that repatriating Japanese managers leave with information that “only they know.” Incorporating “migrating one’s own tasks into the system” as a KPI for Japanese managers drives standardization activities before repatriation at an organizational level.

9. Investment Decisions: Planning a 3-Year ROI for a Knowledge Silo Elimination Project

In investment proposals to Japanese headquarters, an explanation of “to eliminate knowledge silos” alone is rarely sufficient for approval. What is needed is a business case expressed in numbers.

Use the following calculation examples as a reference and apply them to your own situation.

Sample Cost Reduction Calculations

Defect rate improvement: A 20–30% reduction in monthly defective product costs (rework and scrap) is a reasonable benchmark. If current monthly losses total 500,000 yen, the annual reduction would be 1.2–1.8 million yen
Training cost reduction: Shortening the onboarding period from six months to three months reduces veteran trainer hours and accelerates new hire productivity, saving hundreds of thousands of yen per person
Unplanned downtime reduction: Strengthening preventive maintenance to reduce unplanned stoppages from ten to five per year produces significant improvement depending on the loss per incident (production loss + restoration cost)
Administrative labor reduction: Converting the hours spent on aggregating paper daily reports, Excel data entry, and preparing headquarters reports into an hourly cost equivalent

Investment Cost Benchmarks

The implementation cost for a combination of a production monitoring system, paperless tools, and an inventory management system varies significantly depending on the target scale (number of machines, lines, and users). For a mid-sized factory (20–50 machines, approximately 30 users), it is practical to build the business case around a three-year total cost of ownership including both upfront and running costs. TOMAS TECH provides rough estimates and ROI simulations tailored to individual circumstances.

When presenting to headquarters, the path to approval is to use specific numbers rather than expressions like “it will become more convenient” or “DX will advance” — for example: “Annual defect costs will decrease by XX million yen,” “Management labor will be reduced by XX hours per month,” and “Repatriation handover risk will be reduced by XX%.”

10. A Phased Rollout Roadmap — Start Small and Scale Horizontally

Attempting to roll out a knowledge silo elimination project company-wide all at once increases the probability of failure. Shop floor resistance, concentrated implementation costs, and responding to unforeseen issues — large-scale deployments carry higher risk. TOMAS TECH recommends a phased approach, starting with “1 line, 1 process, 1 warehouse, 1 form.”

Phase	Estimated Duration	Key Activities	Expected Outcomes
Phase 1 Visualization	1–3 months	Implement production monitoring on the 1 most problematic line. Digitize 1 type of daily report. Begin recording downtime causes and defect details	Downtime patterns become visible. Data collection habit established. Shop floor resistance identified
Phase 2 Analysis and Improvement	3–6 months	Implement improvements based on accumulated data. Revise and update SOPs. Pilot inventory management system (1 warehouse)	Measurable improvement in defect rate and downtime rate. Data utilization embedded on the shop floor
Phase 3 Horizontal Scaling	6–12 months	Roll out the model from the successful line to other lines. Further digitize forms. Expand inventory management to all warehouses	Factory-wide data foundation established; onboarding period shortened and handover quality improved
Phase 4 Continuous Improvement	12+ months	Explore AI-driven anomaly detection and predictive maintenance. Ongoing activities using data to improve cost and yield	Structural elimination of knowledge silos. Continuous cost reduction and quality stabilization

The key is to position “starting to collect data” in Phase 1 as the first milestone. If expectations of “defect rates will fall immediately” or “costs will drop right away” are too high, the project will be judged as “ineffective” before data has accumulated and will be suspended. The outcome of Phase 1 is that “things that were invisible have become visible,” and sharing that recognition across management, the shop floor, and headquarters is the legitimate basis for moving to the next phase.

11. Failure Patterns and How to Avoid Them — Common Reasons “It Never Takes Hold” on the Shop Floor

Shop floor digitalization and standardization projects frequently end with the outcome that “nobody uses it” or “things revert immediately” after implementation. Below are representative failure patterns and how to avoid them.

Failure Pattern 1: “Top-Down Implementation” Without Involving the Shop Floor

The pattern where the IT department or headquarters selects the system and instructs the shop floor to “use it.” In Thai factories, the dynamic where Japanese managers take the lead and Thai staff become passive is particularly easy to occur. The mitigation is to involve shop floor leaders (Thai staff) from the pilot stage, giving them a sense of ownership — “we chose this system.”

Failure Pattern 2: Delayed Implementation Due to Pursuing a Perfect System

The mindset of “if we’re going to do it, let’s do everything at once” or “let’s implement a perfect system covering all processes” raises the implementation hurdle and leads to project delays and cancellations. Starting with a small scope — 1 line, 1 form, 1 warehouse — and improving as you go using an agile approach is more appropriate.

Failure Pattern 3: Collecting Data but Never Using It

Even after implementing a production monitoring system and getting data flowing, nobody looks at the data or uses it to make decisions. Data only has value when it is used. Deliberately creating “occasions to use data” — such as using data to discuss improvement topics in weekly regular meetings, or incorporating data into monthly production reports — is necessary.

Failure Pattern 4: Veterans Feeling That “I’m Being Made Obsolete”

When veterans perceive knowledge silo elimination efforts as “activities that reduce my value,” you get not just a lack of cooperation but active obstruction. The key is to deliver the positive message that “we are turning veteran knowledge and experience into organizational assets.” By positioning veterans as “knowledge contributors and SOP reviewers,” their motivation to participate in the initiative increases.

12. Investment Strategy Using BOI — Preferential Measures for Automation and IT Investment

Thailand’s BOI (Board of Investment) offers various preferential measures for automation and digitalization investments in manufacturing. Combining system investment for knowledge silo elimination with a BOI application can reduce the effective cost.

BOI’s eligible categories may include not only robots and automation equipment but also IoT, data analytics, and AI-based production management systems and enterprise IT. However, since the applicable conditions and scope of BOI preferential measures vary by business type and investment content, we recommend verifying the latest official information and consulting a BOI-certified consultant for specifics.

Practical points to consider when exploring a BOI application are as follows.

Confirm BOI application eligibility before deciding on the investment (in some cases, retroactive application is not possible)
Consolidate system implementation costs, equipment costs, and personnel costs (including training expenses) as a single BOI application package
Cross-check BOI requirements (employment of Thai staff, technology transfer, export ratio, etc.) against your company’s plan
Begin capturing with the system — before implementation — the data required in application documents (current production capacity, post-improvement targets)

When presenting to Japanese headquarters, adding “the effective cost can be reduced by XX% by leveraging BOI applications” increases the likelihood of approval.

TOMAS TECH’s Perspective: How We Approach On-the-Ground Challenges

TOMAS TECH is headquartered in Bangkok and provides IT/DX solutions to Japanese manufacturers, logistics companies, food producers, and retailers throughout Thailand and ASEAN. “Eliminating knowledge silos” is one of the challenges we hear most frequently from the shop floor.

The following solutions represent areas where we are particularly well-positioned to help.

PEGASUS (Inventory Management System): Manages materials, parts, and product inventory in real time. Inbound/outbound records, lot management, and reorder point alerts eliminate “inventory knowledge silos.” Migrating inventory information previously managed in Excel or Access to a system dramatically reduces the risk of personnel changeovers. Particularly suited to factories that require BOI-compliant traceability, factories operating multiple warehouses, and factories where material substitution and diversion occur frequently.
i-Reporter (Paperless Solution): Digitizes daily reports, inspection records, and quality checklists on tablets and smartphones. Japanese/Thai bilingual input forms reduce the language barrier between Japanese managers and Thai staff. Photo attachment, real-time aggregation, and historical data search convert shop floor records into “organizational knowledge.”
Production Monitoring System: Records and visualizes equipment running, downtime, changeover, and defects in real time. Supports downtime cause analysis, OEE tracking, and preventive maintenance planning. Because “who did what, why the equipment stopped, and how it was restored” is captured as data, equipment management silos are eliminated.
Smartwatch System: Digitally records shop floor workers’ walk-around time, work status, and call response. Makes visible the tasks that veterans “handle entirely on their own,” leading to task delegation and standardization.

TOMAS TECH’s strength is not just system implementation — we support the full journey to operational adoption suited to Thai shop floor realities. We help build an environment where Thai staff can use the system in Thai, and work alongside you to create a structure where local teams continue using the system even after Japanese managers have repatriated.

For those who don’t know where to start, we are available to help from the initial stage of identifying shop floor challenges and selecting the right system. Rather than pushing solutions, we believe that starting with a review of the current state and clarifying priorities leads to implementations with faster ROI.

Summary

“Knowledge silos” cannot be resolved overnight, but with the right approach, steady improvement is achievable. Let us review the key points covered in this article.

Knowledge silos occur in three layers — equipment, quality, and changeover/scheduling — and Thailand’s unique workforce mobility and language barrier make the problem more serious
The nature of “intuition” is a combination of data points, and it can be converted into organizational knowledge through data collection and standardization
Three foundations — production monitoring, paperless records, and inventory management — form the core of knowledge silo elimination
Investment decisions should be built around a business case using the numbers: defect cost reduction, training cost reduction, and administrative labor reduction
Start small — “1 line, 1 form, 1 warehouse” — and scale horizontally while measuring outcomes
Position veterans not as people to be “replaced” but as “knowledge contributors,” thereby earning shop floor cooperation
Explore BOI preferential measures from the planning stage to reduce the effective cost of investment

In Thailand’s 2026 business environment, with ongoing increases in labor costs, logistics costs, and raw material costs, reducing shop floor waste and improving management quality is a management imperative on par with — or exceeding — sales growth. Converting veteran intuition into data and standardized work is one of the most practical approaches to achieving this.

TOMAS TECH provides consistent support from shop floor challenge identification through implementation and operational adoption. We welcome you to start by sharing your current challenges with us. Contact us here.