Textile production management: a guide for industrial directors

In a garment factory near Famalicão, at month's end, the industrial director has three Excel sheets open, a black-covered notebook with the manufacturing orders jotted down by hand, and a WhatsApp from the printing foreman saying that batch 4471 is stuck because nobody knows where the approved colour is. The parent company — one of the fast fashion giants — wants to know whether Wednesday's delivery still holds. The honest answer is "maybe". And "maybe" is not an answer you give to Inditex.

Textile production management is, above all, the elimination of that "maybe". This guide is the reference for Portuguese industrial directors who want to swap the black-covered notebook for a system that tells the truth in real time — and that survives the shop floor, the stubborn foreman and the tax authority's auditor.

Let us be blunt about one thing right at the start: the problem of the Portuguese textile industry is not technological in the sense that software vendors like to paint it. The machines exist. The capital exists — especially since European reshoring put the brands looking again at the Vale do Ave. What is missing is the information layer that connects the parent company's order to the roll of knit coming off the loom. That is the void. And it is a void that costs money every day, without appearing under any accounting heading with the right name.

1. The real operational problem

The Portuguese textile industry does not have a quality problem. It has a visibility problem. The factories of the Vale do Ave produce knit, finishings and garments at a level that justifies European reshoring — but most manage production with instruments that cannot answer the simplest question: where, exactly, is manufacturing order 4471 right now?

Note the precision of the question. It is not "how is production going in general". It is the exact location of a specific order, with its quantity, its approved colour, its promised date. In a factory working for Inditex, Decathlon or Tom Tailor, this question arrives by email at eight in the morning and demands an answer before lunch. Whoever cannot answer it in seconds slowly loses the buyer's trust — and the buyer's trust is the only asset that separates a profitable garment factory from one that survives on residual orders.

The notebook, the Excel and the foreman's memory

In a typical textile factory of 80 employees, the critical knowledge lives in three places: the planning spreadsheet, the foreman's head and the quality control notebook. When the foreman goes on holiday, the factory runs at 70% capacity for a week. This is not hypothesis — it is the pattern.

The most dangerous case is the third repository: the foreman's memory. That man who has been with the firm for 25 years knows, by heart, that customer X tolerates half a shade of colour deviation but customer Y returns the whole batch; he knows that a particular sewing machine tends to skip stitches on heavier fabric and therefore reserves it for light knit; he knows that the subcontracted printer over near Vizela always takes a day longer than it promises. None of this is written down. All of it disappears the day he retires — or, worse, the day he falls ill without warning.

The cost does not appear in the accounts under the name "lack of a system". It appears in disguise: batches redone because of a wrong colour, deadlines that slip and turn into penalties from the parent company, excess raw material stock "just in case", overtime in the last week of the month to plug planning gaps.

Anatomy of a hidden cost: the redone batch

It is worth dissecting a single redone batch, because the exercise reveals how much invisibility costs. An order of 600 pieces in four colours. The printer receives the wrong colour reference — because the technical sheet lived in an out-of-date version of Excel — and produces 150 pieces in the wrong shade before anyone notices. Add it up: the wasted raw material, the labour hours already invested, the machine time that could have produced something else, the delay caused to the rest of the production queue, and — the most expensive of all — the risk of missing the parent company's date and paying a penalty. A single redone batch in a month is rarely dramatic. The problem is that in a factory without a system this happens several times a month, and the annual sum easily reaches the value of an entire digitalisation project.

The three pain points that appear in every audit

• Batch traceability absent or manual. When the brand asks for compliance with the EU Strategy for Sustainable and Circular Textiles, the factory needs to prove where each thread, each dye bath, each finishing batch came from. Without digital traceability, this is archaeology in notebooks.
• Real efficiency unknown. Ask an industrial director what the OEE of their knitting operation is. If the answer is a round number given from memory, it is a guess. Most factories measure neither Takt Time nor real capacity per machine.
• Industrial cost per article estimated, not calculated. The selling price to the parent company is negotiated on the basis of a cost that nobody has validated with shop-floor data. Margins evaporate on articles that everyone swore were profitable.

You cannot manage what you do not measure, and the Portuguese textile industry measures production with the same precision as measuring salt "by eye".

The difference between spinning, dyeing and garment-making

A common error in conversations about the sector is to treat "textile" as a homogeneous block. It is not. Each stage has its own management pathology. In spinning, the problem is the yield of the material — how much yarn comes from how much fibre, at what count, with what waste. In dyeing and finishing, the central problem is the variable bath yield: the same procedure can give different results depending on the raw material batch, the water, the temperature, and so reprocessing is a reality that has to be modelled, not ignored. In garment-making, the problem is the combinatorics of the colour-size matrix and the fragmentation of the operation across dozens of manual stations. A system that serves one stage well may fail utterly at another — and that is why the verticalisation of the software matters so much.

Why generalist ERPs fail here

Textiles have characteristics that generalist ERPs do not model without violence: the colour-size matrix, cutting waste, the variable bath yield in dyeing, the subcontracting of stages (the knit goes to the printer, comes back, goes to finishing). When you force a generalist ERP to manage this, the result is a parallel Excel layer that cancels out the investment. We are back to the black-covered notebook — only now with more expensive licences.

The classic symptom is this: the company buys a respectable generalist ERP, pays for the implementation, pays for the training, and six months later planning still lives in a separate spreadsheet because "the system can't do the matrix". Production never migrates into the ERP. All that stays there is the accounting and the invoicing — which, those, any ERP does. The project money paid for half a system. The half that hurt stayed on paper.

2. What textile production management is exactly

Textile production management is the set of processes, data and systems that turn an order into a finished product with control over deadline, cost, quality and traceability. It is not "having an ERP". It is the ability to plan, execute, capture and reconcile the production of yarns, knits, fabrics, finishings and garments.

The key word in this definition is reconcile. Many systems plan well and execute poorly the verification of what actually happened. Reconciliation — comparing what was planned with what was produced, in quantity, in time, in cost and in quality — is what distinguishes production management from the mere issuing of orders. Without reconciliation, planning is a letter of intent that nobody audits.

The components no one can skip

• Planning (MRP/APS). Calculates raw material needs and capacity from the orders. In textiles, it has to know how to deal with dyeing deadlines, imported yarn lead times and subcontracting windows.
• Manufacturing orders with a matrix. A garment MO is not "500 t-shirts". It is 500 t-shirts in 4 colours × 5 sizes = 20 combinations, each with its quantity, its consumption and its control.
• Shop-floor capture (MES). The PLM defines what to produce; the MES records what was actually produced, when, on which machine, with what efficiency and with what defects.
• Batch traceability. Links each finished product to the dye bath, to the roll of knit, to the yarn supplier. Non-negotiable for sustainability and quality.
• Industrial costing. Aggregates materials, labour and overheads per MO, comparing budgeted against actual.

The terms you will hear in meetings

OEE (Overall Equipment Effectiveness) measures availability × performance × quality. A stopped, slow or defect-producing loom brings the number down. The MES (Manufacturing Execution System) is the execution layer between the ERP and the machine. The MRP (Material Requirements Planning) explodes the bill of materials to generate purchase and production orders. SAF-T is the fiscal file that the tax authority requires. And Kaizen is the discipline of continuous improvement that makes today's OEE the baseline for tomorrow.

Breaking down the OEE: why a figure of 65% does not tell everything

There is a trap in OEE: the aggregate number hides the cause. A knitting operation with an OEE of 65% can have that value for three completely different reasons, which call for three completely different actions. If availability is low, the problem is stoppages — breakdowns, lengthy changeovers, lack of material at the station. If performance is low, the machine is running below nominal speed — mis-tuning, knit that snags, an operator who has not mastered the equipment. If quality is low, a lot of defects are produced that have to be redone. Reporting only the aggregate OEE is like reporting a fever without knowing whether it is flu, an infection or exertion. The value of a good MES lies in breaking down the three factors, per machine and per shift, so that corrective action hits the target.

The aggregate OEE is a diagnosis that does not name the disease. The value lies in knowing whether the number fell because of stoppage, speed or defect.

A brief history — from the shuttle loom to IoT

The textile industry of the North mechanised in the 19th century, automated in the 1980s, and computerised its accounting in the 1990s. But the shop floor stayed decades behind the back office. The great current turning point is automatic capture: sensors on machines, industrial terminals at the stations, code reading on batches. What was paper becomes data — and the data becomes decision.

This time lag explains a great deal. The finance department of a Portuguese textile factory has, for 25 years, worked with systems that give the cash position, the VAT, the trial balance. The shop floor continues, in many cases, with the same information technology of 1985: the operator notes it on paper, the foreman transcribes it to the board, someone types it into Excel at the end of the day — if they type it. The asymmetry between the two worlds is the central problem. The back office knows to the cent how much was invoiced; nobody knows for sure how much it cost to produce.

The back office of the Portuguese textile factory entered the 21st century 25 years ago; the shop floor is entering it now.

3. The picture in Portugal today

The textile and clothing industry (ITV) is one of the pillars of Portuguese exports. According to ATP — the Portuguese Textile and Clothing Association, the sector represents thousands of companies, the overwhelming majority SMEs, and an export volume that places Portugal among the reference European suppliers for the major brands. The cluster is concentrated in the Vale do Ave: Famalicão, Guimarães, Barcelos, Vizela.

The Portuguese competitive positioning has a characteristic that defines the whole digitalisation strategy: Portugal does not compete on price with Asia, and never will. It competes on proximity, on flexibility, on speed of response and on quality. The brands that bring production closer to Europe do so to shorten the cycles — to put a collection in the shop in weeks rather than months — and to reduce the risk of stranded stock. Now, speed and flexibility demand real-time information. A factory that does not know its real capacity cannot promise short deadlines with confidence. Production digitalisation is not, in this context, a luxury of modernisation — it is the condition that makes the commercial argument for reshoring possible.

Business structure: small, family-owned, specialised

According to INE, the overwhelming majority of Portuguese industrial companies are SMEs, and in textiles the average size is particularly small — many garment units operate with 20 to 100 employees. This defines everything: the investment in systems, the decision structure (CEO + CFO + IT lead), and the cultural resistance to change on the shop floor.

The decision trio and the IT hero without a diploma

Whoever sells software to Portuguese industry knows this triangle well. The CEO, usually the owner or the owner's son, decides the strategic matters and wants to see the number that changes. The CFO controls the investment, distrusts anything that does not bring measurable return, and is the one who most values the argument of industrial cost and the execution of incentives. And then there is the IT lead — frequently a self-taught individual with fifteen years in the firm, who has no engineering diploma but knows every interface, every data export, every particularity of the process better than any external consultant. This last one is, simultaneously, the greatest ally and the greatest obstacle of a project. An ally because they know where the bodies are buried; an obstacle because any new system threatens their command of the knowledge. A project that does not treat this IT person as a protagonist — that wants to replace them rather than equip them — is doomed before it begins.

In a family textile factory, the system is not decided by whoever signs the cheque alone; it is decided by the CEO-CFO-IT trio, and the silent veto of the IT person kills more projects than any objection over price.

Productivity: the European challenge

Eurostat data consistently show that productivity per worker in Portuguese manufacturing is below the EU-27 average. In textiles, part of this deficit does not come from the machines — it comes from the organisation of production, from planning and from measurement. A factory with modern equipment and paper-based production management wastes the equipment's advantage.

There is a nuance here that it is important not to oversimplify. The Portuguese productivity deficit is multifactorial — the small size of companies weighs on it, the qualifications structure weighs on it, historical investment weighs on it. But within what an individual factory controls, the organisation of production is the most immediate lever. It does not require buying a new loom; it requires knowing how to make better use of the loom you already have. And that is precisely the part that digital production management attacks: extracting more value from the same asset, reducing stoppages, reprocessing and dead time between stages.

The ageing of the workforce and the urgency of capturing knowledge

A structural fact that rarely enters conversations about software: the workforce of Portuguese textiles and garment-making is ageing, and generational renewal is difficult. Young people do not seek, en masse, the work of a seamstress or a finishing operator. This has a direct consequence for production management: the tacit knowledge accumulated over decades — that of the foreman who knows everything by heart — is approaching retirement without organised transfer. Every year that passes without capturing that knowledge in a system is a year of added risk. Production digitalisation is, seen from this angle, an operation to preserve knowledge before it walks out the door.

Funding: the PT2030 and PRR window

The moment is favourable to investment. PT2030, the PRR, COMPETE 2030 and Norte 2030 fund industrial digitalisation and Industry 4.0. Experience teaches one thing: the applications are approved easily at the intention stage and get stuck at the technical report stage, when it is necessary to prove measurable results. A system that measures OEE and industrial cost is not just operational — it is ammunition for the incentive's execution report.

The error of applying for technology without the indicator

We see this repeatedly: the incentive application describes the equipment and the software to be bought, but is vague about the indicator it commits to improving. Then, at the report stage, the promoter discovers there is no way to prove the improvement because they never measured the starting point. The practical rule is simple and saves a lot of pain: no industrial digitalisation application should proceed without a measured baseline of the indicator it will promise to improve. If you are going to promise to raise the OEE, measure the OEE beforehand. If you are going to promise to reduce reprocessing, count the redone batches beforehand. The honest baseline is not bureaucracy — it is what turns an approved application into an incentive actually paid. IAPMEI and the CCDRs do not pay for intentions; they pay for demonstrated results.

Modern equipment in a factory with paper-based planning is a Ferrari driven in second gear.

4. The implementation models

There are four real approaches to digitalising textile production management. There is no right one — there is the right one for your size, maturity and complexity.

Model A: generalist ERP + parallel Excel

It is the starting point of most. A financial and invoicing ERP, certified by the tax authority, with production managed in separate spreadsheets. It works for the accounting. It fails at production: the shop-floor data never reconciles with the accounting, and the industrial cost is fiction.

It is important to recognise that this model is not stupid — it is merely limited. For a very small garment factory, fully subcontracted, that receives cut pieces and returns sewn garments, the parallel Excel may be sufficient for years. The problem arises when the company grows, gains its own stages, starts buying raw material directly, and the complexity exceeds the spreadsheet's ability to represent it without errors. The sign that this point has been reached is when the planning Excel comes to have more than one person editing it, and the versions start to diverge.

Model B: integrated vertical textile ERP

An ERP designed for industry, with colour-size matrix, subcontracting management, bath yield and costing per MO native. The MULTI ERP fits here: modular, with production, purchasing, sales and planning modules designed for textiles, footwear and clothing. The advantage is not needing a parallel Excel. The data is born right.

The subtle gain of this model is the uniqueness of the information. When the article is defined once, with its matrix, its bill of materials and its routing, and that definition feeds planning, purchasing, production and invoicing, the work — and the error — of re-entering data disappears. The technical sheet the commercial department sees is the same one the shop floor uses. This coherence eliminates an entire class of errors: those born of two versions of the truth.

Model C: vertical ERP + MES capture layer

The ERP plans and costs; a real-time capture layer records what happens on the machines. KORA Productivity does this: industrial terminals at the station, production capture, OEE calculation and operational efficiency. It is the model that turns guesswork into measurement. Indispensable for factories that want to compete on deadline and on cost.

This is the qualitative leap. In models A and B, what happened on the shop floor reaches the system with delay and with the unreliability of someone filling in paper at the end of the shift. In model C, the data enters at the moment and in the place where the event occurs. The operator finishes an operation and records it on the terminal; the stoppage is recorded with its cause; the defect is counted when it appears. The OEE ceases to be an end-of-month calculation to become a living indicator, which the foreman consults in real time and uses to react within the shift itself.

Model D: high-complexity low-code ERP

For groups with multiple units, very specific processes and a need for continuous adaptation, QAD Adaptive ERP offers a low-code engine that moulds itself to complex processes without rewriting code. It is the option for the complexity the other models cannot accommodate — typically groups above 150-200 employees or with multi-country operations.

The advantage of low-code in a large group is the speed of adaptation. Processes change — a new stage comes in, a unit opens, an international customer's requirement changes — and the ability to reconfigure the system without a long development cycle becomes a strategic asset. For a small SME, this flexibility is over-engineering: you pay for capacity you do not use. Hence the importance of calibrating the model to the real size, and not to the ambition.

The trade-off no one tells you about

The true cost of any model does not lie in the licence. It lies in the adoption. The printing foreman with 25 years in the firm who hides the capture tablet with tape behind the machine is not a villain — he is a sign that the project did not involve the shop floor. The technology is the easy part. Changing the habit is the expensive part.

The engineering of adoption has rules we have learned to respect. The capture has to be faster than the paper it replaces — if it takes longer to record on the terminal than to scribble in the notebook, the operator sabotages it, and he is right. The terminal has to be within arm's reach, not twenty metres from the station. The information the operator enters has to come back to him usefully — if he only feeds the system and never sees anything in return, he feels watched, not helped. And the foreman has to be the first to use and defend the system, because the team follows the foreman, not the consultant.

5. How to assess whether your company needs it

Not every factory needs an MES tomorrow. But every factory needs to know where it stands. This diagnosis separates urgency from technological vanity.

Signs that you are losing money for lack of a system

• The answer to "what was last week's OEE?" takes more than five minutes or is a guess.
• Batches are redone because of wrong colours or mixed-up specifications more than once a month.
• The industrial cost of an article has never been reconciled with real production data.
• When the parent company asks for batch traceability, someone goes looking through notebooks.
• Overtime is concentrated in the last week of the month to plug planning gaps.
• Planning knowledge lives in a single person, and that person cannot fall ill.

Digital maturity in four tiers

Before choosing a solution, it is useful to place the company on a maturity tier, because the right leap is always to the next tier, never two at once. At tier 1, production is managed by paper and memory; there is no data, there is intuition. At tier 2, there is an ERP that does the accounting and the invoicing, but production lives in a parallel Excel; there is financial data, there is no industrial data. At tier 3, the vertical ERP integrates production and the data is born inside the system, although the shop-floor capture is still manual. At tier 4, the capture is automatic and real-time, the OEE is live, the industrial cost is actual, and BI closes the decision loop. Trying to leap from tier 1 straight to tier 4 is the recipe for failure — the organisation does not absorb the change. The path is successive leaps, each consolidated before the next.

Step by step: the diagnosis in five stages

1. Measure your current OEE for two weeks, even by hand. Note, per machine and shift, available time, productive time, good pieces and defective pieces. The number that comes out — however ugly it is — is your baseline. Without a baseline there is no measurable improvement.
2. Trace the journey of a manufacturing order from start to finish. Choose a real MO and follow it physically: where it stops, how long it waits between stages, how many times it changes hands, where information is lost. Each stoppage is a hidden cost.
3. Calculate the real cost of a "profitable" article. Take an article that everyone swears is profitable. Add up real materials, real labour and overheads. Compare with the selling price. The surprise is usually unpleasant and instructive.
4. Identify where the critical knowledge lives. List the processes that stop if a specific person is absent. Each of those is an operational risk that a system mitigates.
5. Quantify a month's waste. Redone batches, deadline penalties, idle stock, avoidable overtime. Add it up. That number is your potential ROI — and the argument for the CFO.

Before buying software, measure the OEE with pencil and paper for two weeks. The honest baseline is worth more than any demo.

Quick wins before any project

• Standardise the production technical sheet. A clear sheet, with coded colours and defined tolerances, eliminates half the redone batches. It is implemented in a week, without new software.
• Create a visual machine-load board. A simple visual management board on the shop floor gives immediate visibility and prepares the culture for digital capture.

What NOT to do in the diagnosis

There are diagnostic errors that compromise the entire subsequent project. Do not start with the technology — start with the pain. The correct question is not "what system should I buy?", it is "where am I losing the most money and why?". Do not measure everything at once; choose the bottleneck and measure it well, because a shallow measurement of the whole factory is worth less than a deep measurement of the critical point. Do not trust the number the foreman gives off the top of his head — he believes it, but it is a guess, and the guess is exactly what we came to eliminate. And do not confuse activity with productivity: a machine busy producing what there is no order for is not efficiency, it is stock piling up.

6. What to choose and why

The right decision depends on size and complexity. Here is the matrix we use with Portuguese industrial directors.

Decision matrix by size

Why real-time capture is the watershed

The difference between a factory that knows and one that guesses does not lie in the ERP — it lies in the capture. KORA Productivity records production at the moment it happens, at the station, without the operator having to fill in paper at the end of the shift (which nobody fills in with accuracy). It is the layer that makes the OEE real and the industrial cost defensible before the CFO and before the PT2030 incentive auditor.

The parallel with footwear: SKU complexity as a stress test

Whoever doubts the need for vertical software should look at footwear from Felgueiras. A sample collection easily has between 800 and 1200 SKUs, organised across three axes — colour, size and fitting — that multiply into a combinatorics that no generalist ERP models without contorting itself. International buyers visit twice a year, in tight windows (men's footwear in August, women's in February), and the factory has to have the collection ready and costed for those moments. This complexity is the ultimate stress test of a vertical system: if it withstands footwear's tri-axial matrix, it withstands the colour-size matrix of textiles with ease. Textiles and clothing share the same structural demand, albeit with fewer axes.

Do not buy BI before you have data

A common error: buying beautiful dashboards before having reliable data to feed them. Qlik Sense turns production data into decision — but only if the data exists and is reliable. First the capture, then the BI. The reverse order produces expensive panels that show rubbish elegantly. It is worth reading our guide to Business Intelligence in the textile industry to understand the correct sequence.

Buying dashboards before having reliable capture is putting a panoramic window in a wall with no house behind it.

Integration between subsidiaries and subcontractors

Whoever subcontracts stages — and in Portuguese textiles almost everyone subcontracts — needs to link their ERP to the partners' operations. MULTI Connect links subsidiaries and partners of the MULTI ERP, giving visibility over the stage that is off-site. The printer who takes a day longer stops being a Thursday surprise.

The warehouse and dispatch: do not forget the end of the line

Production ends but the cycle does not. The finished product has to be stored, separated by order and dispatched without error to the parent company — and a dispatch error (mixed-up colours, wrong quantities on a packing list) destroys the trust earned in months of well-done production. KORA Inventory Suite handles warehouse management, picking and the packing list with the same traceability rigour as production. It is worth recalling a shop-floor truth: the warehouse manager of a distribution centre in the Lousada-Paços corridor does not let go of the radio for more than two hours, and any rollout that takes him off the operation for too long faces fierce resistance. The introduction of systems in the warehouse is done in short, phased windows, respecting the rhythm of dispatch.

7. Applicable regulatory framework and compliance

Production management does not live isolated from the law. Four regulatory fronts touch the Portuguese textile production system directly.

Invoicing and taxation: DL 28/2019 and SAF-T

Decree-Law 28/2019 requires invoicing software certified by the tax authority, with ATCUD on each document. Ordinance 195/2020 mandates the monthly reporting of the SAF-T. Any system that issues fiscal documents — including production notes that turn into invoices — has to comply. A serious vertical ERP is born certified; a parallel Excel that issues documents is a fiscal time bomb.

The point that escapes many directors is that the fiscal obligation is not limited to the final invoice. Transport notes, delivery notes to subcontractors, movement documents — all of this enters the perimeter of certification and reporting. In an operation with extensive subcontracting, where the knit travels to the printer and comes back, the quantity of circulation documents is considerable, and each one has to be compliant. Managing this outside a certified system is accumulating fiscal risk that, sooner or later, an inspection will find.

Sustainability: the EU Strategy for Sustainable and Circular Textiles

The European strategy for sustainable and circular textiles will require a digital product passport and traceability along the chain. The major brands are already transferring this requirement to Portuguese suppliers. Without digital batch traceability, the factory falls out of the specifications of the parent companies that pay the most. This is not optional compliance — it is a condition of access to the premium market. Our guide on digitalising textile collections and technical sheets delves into collaboration along the chain.

It is worth being realistic about the timetable and the pressure. The digital product passport is a requirement that materialises in phases over the coming years, but the leading brands do not wait for the letter of the law — they anticipate it in their specifications, because sustainability is today a selling argument to the end consumer. The Portuguese supplier who manages, today, to generate an end-to-end traceability report from the system has a concrete commercial advantage over the competitor still working in notebooks. Future compliance, anticipated, becomes present differentiation.

Batch traceability has ceased to be a requirement of internal quality and become the ticket of entry into the specifications of the brands that pay best.

Cybersecurity: NIS2 and the connected factory

Directive (EU) 2022/2555 (NIS2), transposed into national law, extends cybersecurity obligations to more sectors and to their respective supply chains. A factory with production capture, industrial terminals and integration with parent companies is, by definition, an attack surface. Immutable backup, network segmentation and VPN for remote access cease to be a luxury. The guide to NIS2 technical controls for factories and the article on immutable backup for industrial SMEs set out the practical path.

There is a dimension of NIS2 that catches SMEs by surprise: the supply chain effect. Even a factory that is not itself an entity covered by the directive may be obliged to comply with security requirements by contract, because its parent company — which is covered — transfers the requirements to its suppliers. That is, information security becomes, like sustainability, a contractual condition of supply to the major brands. The factory that digitalises production and integrates with customers has to do so with security designed from the start, not added hastily when the customer asks for the cybersecurity questionnaire.

Personal data and document management

The GDPR and Law 58/2019 apply to employee data captured on the shop floor (productivity per operator, for example). The ISO 27001 standard structures information security. And documentary digitalisation — technical sheets, certificates, approvals — benefits from a qualified eIDAS signature. Document Management eliminates paper while maintaining the legal archive; our guide to document management integrated into the ERP shows how.

The capture of productivity per operator is a delicate terrain that deserves care. Measuring how much each operator produces has legitimate management value, but it directly crosses the GDPR and respect for dignity at work. The practical rule: the capture should serve the improvement of the process, not the surveillance of the person. When individual productivity data are presented as an organisational tool — to balance loads, identify training needs, recognise performance — acceptance is much higher than when they are experienced as a surveillance camera. The difference lies in transparency: informing employees of what is measured, why, and how the data are used. Law 93/2021, on the whistleblowing channel for companies with 50 or more employees, is a further reminder that the relationship with workers today has a layer of compliance that did not exist before.

8. How INFOS approaches this

We have worked for more than three decades with Portuguese industry — textiles, clothing and footwear included. Our position is simple: production software has to model the reality of the sector, not force the sector to contort itself to fit the software. That is why the MULTI ERP has colour-size matrix, subcontracting and costing per MO native — not as an improvised module, but as a backbone.

On shop-floor capture, the approach is phased and honest: we do not impose terminals at every station on the first day. We start with the stages where the lack of data hurts most — typically the factory bottleneck — and expand with KORA Productivity as the culture adheres. Shop-floor adoption is the success factor we respect most, because it is where projects die.

What we learned from projects that went wrong

Honesty compels: not every project goes the way the slide deck promises. We have learned, at our own cost and at that of the clients, some lessons that today shape the way we work. We learned that imposing capture across the whole factory at once generates wholesale rejection — so we phase it. We learned that a system configured without the in-house IT person is never truly adopted — so we treat them as a protagonist, not an obstacle. We learned that promising an unrealistic implementation deadline to close the deal destroys trust when the deadline slips — so we prefer an honest schedule to an optimistic one. And we learned that terminal training is not a one-day event, it is weeks of accompaniment, with a presence on the shop floor in the first phase. Technology is rarely the cause of failure; change management almost always is.

Production software has to model the reality of the sector, not force the sector to contort itself to fit the software.

The ecosystem around production

Frequently asked questions

How can I know in real time where a specific manufacturing order is?

An integrated production management system allows each order to be tracked from entry to dispatch. Without this, you depend on emails, Excel and the foreman's memory. Real-time visibility eliminates the "maybe" in answers to customers and reduces delays caused by orders lost in the process.

What is the real cost of a batch redone due to a colour error?

A redone batch includes wasted raw material, invested labour, lost machine time, delay in the production queue and the risk of a penalty from the parent company. In a factory without a system, this happens several times a month. The annual sum frequently equals the investment in a digitalisation solution.

What is batch traceability and why does it matter for audits?

Digital traceability proves the origin of each thread, dye bath and finishing batch. Brands require this for compliance with sustainability regulations. Without a system, it is archaeology in notebooks. Manual traceability is slow, imprecise and does not survive rigorous audits.

How do I calculate the real OEE of my knitting operation?

OEE (Overall Equipment Effectiveness) measures availability, performance and quality. It requires continuous data on Takt Time and real capacity per machine. Without digital measurement, most industrial directors give round numbers that are guesses. Measuring correctly reveals hidden inefficiencies.

Why is my industrial cost per article estimated and not calculated?

Prices are negotiated with customers on the basis of costs that nobody has validated with real shop-floor data. This causes margins that evaporate on articles that seemed profitable. An integrated system calculates precise costs per article, enabling more secure negotiations.

What is the difference between managing spinning, dyeing and garment-making?

Spinning: material yield (how much yarn comes from how much fibre). Dyeing: variable bath yield (frequent reprocessing). Garment-making: colour-size combinatorics and fragmentation across dozens of manual stations. Each stage requires different metrics and control.

What happens when the foreman goes on holiday or retires?

The factory runs at 70% capacity. The critical knowledge lives in his head: customer tolerances, machine characteristics, real subcontractor deadlines. Nothing is documented. A management system captures this knowledge and makes it accessible to the whole team.

Sources

• Regulation (EU) 2023/1115 — Corporate sustainability due diligence (CSDDD), applicable to textile companies with an international supply chain
• EU Strategy for Sustainable and Circular Textiles (2022) — European Commission, Directorate-General for the Environment
• ISO 22301:2019 standard — Business continuity management systems, applicable to industrial planning and traceability
• National Statistics Institute (INE) — Statistics of the Portuguese Textile Industry and Operational Efficiency Indicators
• IAPMEI — Guide to Good Practice in Production Management for Portuguese Industrial SMEs