Predictive Maintenance for Pulp and Paper Mills: Predictive Maintenance Indonesia

Direct answer (AEO): Pulp and paper mills concentrate two very different risk profiles under one roof, and a predictive maintenance programme must handle both. The fibre line — digester, blow line, washing, screening, bleaching — is a continuous chemical process where an unplanned stop drains the whole system and can force expensive re-start procedures. The paper machine is a high-speed mechanical system where a single bearing or roll failure destroys a four-figure-cost fabric and takes days to recover, and where product quality is directly coupled to machine condition. On top of both sits the recovery boiler, which is simultaneously the mill’s steam supply, its chemical recovery unit, and its single greatest safety exposure. In Indonesia, where pulp and paper is a major export industry, a programme concentrated on the paper machine and the recovery boiler typically protects the majority of the mill’s unplanned-loss exposure.

Running a pulp or paper mill in Indonesia? Contact Tiaravib — email info@tiaravib.com or WhatsApp +62 821-1378-3727.

Why the Paper Machine Is the Financial Centre of Gravity

Paper machines are among the most expensive pieces of equipment in any manufacturing environment per metre of length, and their failure modes are unusually destructive in financial terms. A machine running at a thousand metres per minute will consume and destroy forming fabric, press felt, and dryer clothing if it stops abruptly or if a component fails in a way that damages the sheet path. The clothing set for a single machine can cost more than a high-value rotating machine, and unlike that machine it cannot be repaired — it is a consumable that must be replaced and re-threaded.

Beyond the clothing, the recovery time is the dominant cost. Restarting a paper machine is not a matter of pressing a button: the stock system must be stabilised, the sheet re-threaded, and the machine brought back to specification grades, a process measured in hours even in the best case and days when damage is involved. Every hour lost is production at paper-mill margin, and every restart consumes clothing life.

The failure modes that create this exposure are well understood and all of them are detectable. Dryer section bearings degrade steadily and give clear warning. Uhle boxes and suction rolls develop vibration signatures long before surface damage becomes visible on the sheet. Gearboxes and drive trains show tooth degradation and lubrication breakdown in oil analysis and vibration spectra. Calendar and reel drums lose balance over time. Felt and fabric problems produce pressure pulsation signatures that appear in the bearing spectra before they produce a visible defect. The critical point is that all of these progress over weeks, which means a monitoring programme with a two-to-four week interval on the paper machine will catch them — provided the interval is actually maintained and the findings are acted upon.

The Recovery Boiler: Safety and Steam in One Asset

The recovery boiler occupies a category of its own. It burns black liquor to recover pulping chemicals and generate steam, which makes it the mill’s chemical recovery unit and its steam supply at the same time. There is no way to run the mill without it, and there is no way to operate it without accepting a genuine safety exposure, because a black liquor recovery boiler that fails catastrophically can release molten smelt into the lower furnace with explosive consequences.

This means the recovery boiler’s reliability programme sits under a different set of constraints than a normal pressure vessel. Inspection and condition assessment are driven by statutory requirements and industry practice rather than by economic optimisation alone, and the operating regime is bounded by safety mechanisms — smelt bed stability, water circulation integrity, tube wall thickness, and furnace gas flow — that must be monitored continuously rather than periodically sampled.

For a predictive programme, the practical implication is that the recovery boiler is managed primarily by inspection, thickness measurement, and process parameter monitoring rather than by the vibration and oil techniques that dominate on the paper machine. What both share is the need for a coherent programme that knows which asset is being protected by which technique. A mill that applies paper-machine techniques to the recovery boiler, or applies inspection thinking to the paper machine, will underperform on both. The organising logic for that split is what a formal predictive maintenance framework provides.

When Product Quality Is a Condition Signal

One characteristic distinguishes pulp and paper from most other industries: the product quality measurement system is available in-line, continuously, and at high resolution. Caliper, basis weight, moisture, formation, gloss, and profile are all measured while the machine runs. Many mills treat these as process data and separately treat condition monitoring as maintenance data, and thereby miss the most valuable signal available.

Quality variation is frequently the earliest visible symptom of mechanical degradation. A slowly developing bearing fault on a dryer roll produces periodic caliper variation at the bearing’s rotation frequency. A failing uhle box produces moisture profile instability. Drive train degradation produces tension variation that shows up in web breaks. Cross-direction profile problems often trace to roll condition rather than to the headbox or control system. In each case, the quality trend leads the mechanical failure by a substantial margin, and it is visible to the production team before it is visible to maintenance.

Connecting the two data sets is a straightforward engineering exercise with a high return: plot quality parameters against machine speed and against rotation frequencies, and correlate excursions with known component frequencies. Mills that do this routinely identify failing components weeks before the vibration route would have flagged them, because the quality system samples far more frequently than a manual route ever could. This is a genuinely distinctive opportunity in pulp and paper, and it is one of the reasons online monitoring architecture decisions in paper mills often look different from those in other process industries.

Pump, Fan, and Refiner Economics in a Wet Process

A pulp mill moves enormous volumes of liquid and generates substantial airflow, and the associated pumps, fans, and refiners consume a large share of the mill’s electrical load. This creates a reliability argument that does not exist in most other industries: efficiency degradation on these machines is measurable as a direct operating cost, not merely as an availability risk.

A refiner running with worn plates consumes significantly more energy per tonne of pulp, and the degradation is gradual enough that nobody notices it without a measurement. A pump operating away from its best efficiency point, or with a worn impeller, draws excess power continuously. A fan with fouled blades or a damaged bearing moves less air per kilowatt-hour. Individually each of these is a modest figure; across a mill with hundreds of such machines, the aggregate is a material operating cost that sits in the monthly electricity bill.

This is where pulp and paper reliability work intersects directly with lifecycle cost management. The corrective work — replacing worn refiner plates, trimming an impeller, cleaning a fan, restoring a pump to its curve — is cheap and requires no capital approval. The barrier is detection, and detection requires measurement. A mill that measures pump performance quarterly and refiner plate condition against production targets will find recurring savings that pay for the monitoring programme many times over, purely from energy.

Programme Design for an Indonesian Mill

Most Indonesian pulp and paper mills operate with a long-established maintenance organisation and an existing condition monitoring capability of some kind, frequently a portable vibration route and a lubrication programme. The realistic question is not whether to start but how to reallocate effort from where it currently goes toward where the loss actually occurs.

The sequencing that works in practice starts with the paper machine, because that is where mechanical failure translates most directly and most expensively into lost production, and because the monitoring techniques required are mature and well suited to a routine route. It then addresses the recovery boiler through the existing inspection and statutory programme, ensuring that condition and process data are available to the people making operating decisions rather than only to the inspection contractor. The fibre line comes next, focused on the large pumps and agitators whose failure interrupts the pulp supply. Utilities and the balance of the mill follow, largely through oil analysis and thermography, which are cheap per point and highly effective at scale.

The recurring failure in Indonesian mills, as elsewhere, is interval drift. A paper machine route established at two weeks quietly becomes four weeks, then monthly, as the team absorbs other work. At a monthly interval, several of the failure modes described above progress from detectable to destructive between visits. The discipline that prevents this is contractual or structural rather than motivational: either the route interval is written into a service agreement with a defined reporting date, or it is resourced with a named person whose only job is the route. Wishing the team were more disciplined does not work, and every mill that has tried it has watched the interval drift.

Related practice for Indonesian operators is set out in fertilizer and ammonia plant reliability.

Key Takeaways

• Pulp and paper mills combine a continuous chemical process with a high-speed mechanical machine, and the two require different monitoring approaches on the same site.
• The paper machine is the financial centre of gravity: a single bearing or roll failure can destroy clothing and cost days of production.
• The recovery boiler is a safety-critical pressure vessel and the steam source simultaneously, and is managed by inspection and process parameters rather than vibration.
• In-line quality measurements are genuine condition signals, and correlating them with component frequencies frequently detects faults before the vibration route does.
• Refiners, pumps, and fans represent measurable and continuous energy waste when degraded, making performance monitoring a direct operating-cost saving.
• Interval drift is the most common programme failure; the fix is a contractual or structural commitment to the route, not more exhortation.

Frequently Asked Questions

Should the paper machine be monitored more often than the rest of the mill?

Yes. A two-to-four week interval is appropriate on the paper machine because its failure modes progress to destructive within roughly that window, and because the financial consequence of a failure is so much larger. The fibre line and utilities can generally be covered at longer intervals with oil analysis and thermography filling the gaps.

Why is the recovery boiler treated differently from other assets?

Because it combines three roles — pressure vessel, chemical recovery unit, and steam supply — and carries a genuine safety exposure that ordinary assets do not. Its condition is managed through inspection, thickness measurement, and process parameters under statutory requirements, rather than through vibration and oil techniques.

Can quality data really detect mechanical faults?

Frequently yes. Caliper, moisture, and profile variation often appear at frequencies that match specific rotating components, and the in-line quality system samples far more often than any manual route. Correlating quality excursions with component frequencies is one of the highest-return exercises available in a paper mill.

How much does energy waste from degraded pumps and refiners cost?

It varies with mill size, but it is material. Worn refiner plates and pumps operating off their curve consume excess power continuously across the whole mill, and the aggregate appears as a persistent electricity cost that nobody questions because it is not separated from the baseline. Measuring performance periodically is the only way to see it.

What is the most common reason pulp mill programmes fail?

Interval drift. Routes established at two weeks quietly become monthly as the team absorbs other work, and at monthly intervals several failure modes progress from detectable to destructive between visits. The reliable fix is a contractual or structural commitment rather than an appeal to discipline.

Do we need online monitoring on a paper machine?

Not necessarily on the whole machine, but strongly worth considering on the highest-consequence drivetrains. Online coverage captures transient events during grade changes and speed changes that a portable route never sees. Many mills run portable coverage on the dryer section and online coverage on the main drives and gearboxes.

To discuss a reliability programme for a pulp or paper mill, contact Tiaravib at info@tiaravib.com or WhatsApp +62 821-1378-3727.

Pulp and paper rewards a programme that respects the difference between its assets. Tiaravib has worked with Indonesian heavy industry since 1995, and the pattern in mills is consistent: the machines that stop production are few and identifiable, the techniques that protect them are mature, and the programme succeeds or fails on whether the route is actually executed at the stated interval. Standards relevant to this work include ISO 17359 for condition monitoring programme structure, ISO 20816-3 for vibration severity assessment, and ASME codes for pressure equipment inspection practice.