Industrial maintenance becomes significantly more challenging when production equipment operates almost continuously and only limited time is available for planned maintenance.In 2019, I joined Zangak Publishing House in Armenia as Production Manager shortly before the company's annual peak production period. From late May through August, the facility was required to produce large volumes of state-commissioned school textbooks before the beginning of the academic year. During this period, offset printing, folding, stitching, binding, and related equipment could operate for up to 22 hours per day.The challenge was not simply to repair equipment faster after a failure. The objective was to identify developing mechanical problems early enough to prevent them from becoming significant production interruptions.A structured maintenance and production-management approach was implemented using daily inspections, operator feedback, vibration and temperature monitoring, recording of minor stoppages, preventive maintenance, and dynamic adjustment of production sequences.According to the company's internal performance evaluation, overall operational performance during the period improved by approximately 18–20%, while equipment-related stoppages and failures during peak production were reduced by approximately 15–20%. The company completed its textbook production and distribution requirements by August 16, 2019.This case study examines the engineering principles behind that approach and discusses lessons that may be applicable to other manufacturing environments operating under intensive production loads.Disclosure: I worked at Zangak Publishing House as Production Manager from April to November 2019. I have no current financial relationship with the company.1. Operational ContextZangak Publishing House, established in 1997, operates in publishing and industrial printing in Armenia.One of the most demanding periods of its annual production cycle occurs between late May and August, when large quantities of school textbooks must be produced and distributed before the beginning of the academic year.During the peak-production period, the facility relied on multiple interconnected processes, including:offset printing;folding;stitching;binding;material handling;supporting mechanical systems.Some production equipment operated for as much as 22 hours per day.This operating environment created a reliability problem that differed significantly from normal production.When equipment operates for only part of a day, maintenance personnel normally have greater flexibility to inspect, adjust, lubricate, repair, or service machinery outside active production hours.When production approaches continuous operation, those maintenance windows become much smaller.Before the 2019 season, the company had repeatedly experienced difficulties meeting its peak-production deadlines, with delays ranging from approximately three weeks to as much as one and a half months.The maintenance problem and the production problem were therefore closely connected.Equipment reliability was not simply a technical issue. It directly affected whether the company could complete its production commitments on schedule.Maintenance-management literature has long recognized that effective maintenance involves more than repair activity alone and includes maintenance techniques, planning, scheduling, information management, and organizational decision-making [7].2. The Engineering ProblemTraditional reactive maintenance becomes particularly risky under sustained high production loads.Waiting for equipment to fail before intervening may create several consequences:production stops unexpectedly;maintenance personnel must troubleshoot under severe time pressure;required spare parts may not be immediately available;the failure of one machine may affect subsequent production processes;lost production capacity becomes increasingly difficult to recover.Preventive maintenance can reduce these risks, but a purely calendar-based maintenance strategy has its own limitations.The same machine operating eight hours per day and operating close to 22 hours per day is exposed to substantially different operating conditions and accumulated mechanical loading.Condition-based maintenance literature addresses this problem by using information about the actual condition of equipment to support maintenance decisions rather than relying exclusively on predetermined intervals. Jardine, Lin, and Banjevic describe condition-based maintenance as a process involving condition-data acquisition, data processing, and maintenance decision-making [1]. ISO 17359 similarly provides a general framework for establishing machine condition-monitoring programs [2].The central practical question in our case was therefore:Could developing equipment problems be recognized early enough to create a controlled maintenance opportunity before production was forced to stop?3. Practical MethodologyThe approach implemented during the 2019 peak-production period combined several maintenance and production practices into a single operating process.None of the individual techniques was intended to be presented as a new scientific discovery.The practical value came from connecting the information produced by different activities and using it to make earlier maintenance and production decisions.3.1 Structured Daily InspectionsOne of the first steps was to introduce more structured daily inspection routines and adapt existing maintenance checks to the demands of peak production.The inspections were not intended merely to confirm that a machine was running.The objective was to identify changes in equipment behavior before they developed into significant failures.Maintenance decisions before each shift took into account:scheduled maintenance requirements;recent machine behavior;previous interruptions;operator observations;current production intensity.This allowed maintenance attention to be directed toward equipment showing signs of increased technical risk rather than treating every machine as though its condition were identical.This concept is consistent with the broader logic of condition-based maintenance, in which observed equipment condition contributes to maintenance decision-making [1,2].3.2 Operator Feedback as a Maintenance InputMachine operators interact continuously with production equipment.As a result, they can sometimes notice behavioral changes before those changes become obvious during a formal maintenance inspection.These observations may include:unusual sound;increased vibration;changes in machine response;repeated adjustment requirements;irregular movement;abnormal heat;small recurring interruptions.During the peak-production period, operator feedback was incorporated more systematically into maintenance decisions.Instead of considering these observations merely informal comments, they were evaluated together with inspection findings and operating history.In practical terms, the operator became an additional source of equipment-condition information.This principle is consistent with the broader philosophy of Total Productive Maintenance, which places significant emphasis on participation across production and maintenance functions rather than treating equipment care as the responsibility of a completely isolated maintenance department [4].Ahuja and Khamba's review of Total Productive Maintenance literature identifies maintenance practices, organizational implementation, and broader manufacturing competencies as central elements of effective TPM programs [4].3.3 Vibration MonitoringVibration received particular attention on high-speed printing and folding equipment.Rotating mechanical systems can change their vibration behavior as their mechanical condition changes.Vibration analysis is an established tool in machinery diagnostics. For example, Randall and Antoni describe the use of vibration-based signal analysis for diagnosing faults in rolling-element bearings and explain how mechanical defects can produce identifiable changes in vibration characteristics [3].Our application was much more practical and less sophisticated than a laboratory-level vibration diagnostic program.The objective was not to perform advanced spectral analysis on every machine.Instead, the practical question was:Has the vibration behavior of this machine changed compared with what we normally observe?A noticeable increase or unusual change could justify closer inspection.Possible causes could include:looseness;alignment issues;bearing deterioration;imbalance;wear;other developing mechanical conditions.The value was therefore not simply in recording a vibration value.It was in detecting a change relative to the machine's normal operating behavior and using that change as a reason for further investigation.3.4 Temperature MonitoringTemperature was another practical condition indicator used during the production period.Motors, gearboxes, and other important components operating under sustained load were checked for unusual temperature behavior.The same principle applied as with vibration.A component does not necessarily fail at the moment its operating temperature begins to change.A machine may continue producing while a developing mechanical or electrical issue causes a component to run progressively hotter than normal.Recognizing that change before shutdown provides an opportunity to investigate during a controlled maintenance window.ISO 17359 provides the general framework for machine condition monitoring, while the broader condition-monitoring literature emphasizes acquiring and interpreting information about equipment condition before making maintenance decisions [1,2].In our case, temperature was therefore treated primarily as an early-warning signal, not as an isolated maintenance criterion.3.5 Recording Minor StoppagesAnother important change was recording even relatively small equipment interruptions and operational irregularities.Minor stoppages are easy to dismiss.A machine may stop for several minutes.An operator adjusts something.Production restarts.The event may never appear in a formal breakdown report.Individually, such an interruption may seem unimportant.Repeated interruptions, however, can reveal an emerging pattern.Recording these events allowed us to investigate questions such as:Is the same machine stopping repeatedly?Does the problem occur after prolonged operation?Does the interruption appear during a particular type of production?Did operators notice noise, heat, vibration, or unusual adjustment requirements beforehand?Is the frequency of the interruption increasing?This changed the value of small production events.Instead of disappearing after production restarted, they became part of the information used to understand equipment behavior.Condition-based maintenance research similarly emphasizes that maintenance decisions depend not simply on collecting information, but on processing observations into information useful for diagnosis and decision-making [1].3.6 Dynamic Production AdjustmentOne of the most useful parts of the approach involved connecting equipment-condition information with production planning.If a machine began showing early signs of mechanical stress, shutting it down immediately was not always necessary.At the same time, continuing to operate it under the same load until it failed could create unnecessary risk.Where production requirements allowed, the sequence of work could instead be adjusted.For example, jobs imposing lower mechanical stress on affected equipment could temporarily be prioritized.This allowed production to continue while maintenance personnel prepared a targeted intervention.The required repair, adjustment, inspection, or service could then be performed between production shifts rather than after an uncontrolled mid-shift failure.This led to one of the most important conclusions from the project:Production scheduling can itself become part of a reliability strategy.Traditional production planning considers factors such as:customer deadlines;workforce;material availability;machine capacity;production sequence.Under intensive operating conditions, I believe equipment condition should also influence that decision process.The idea of connecting production decisions with maintenance planning is also supported by reliability research. Nourelfath, Fitouhi, and Machani examined integrated production and preventive-maintenance planning rather than treating the two decisions as unrelated problems [6].The industrial application described here was much simpler than such mathematical optimization models, but the underlying principle was similar: production and maintenance decisions interact.4. Integrating the InformationThe most important part of the approach was not any one technique.It was the information flow between them.In simplified form, the process worked as follows:Operator observation↓Daily inspection↓Vibration and temperature observations↓Minor-stoppage history↓Assessment of equipment condition↓Production adjustment when appropriate↓Targeted maintenance interventionThe objective was to create a feedback loop between production and maintenance personnel.Instead of maintenance responding only after a breakdown, the system attempted to identify deterioration earlier and create a planned opportunity to intervene.This distinction is important.Modern condition-based maintenance can involve advanced sensing, prognostics, algorithms, and data-processing technologies [1].However, the underlying logic can also be applied at a more practical level:observe → compare → identify abnormal behavior → assess risk → act before failure.5. ResultsBy August 16, 2019, Zangak Publishing House had completed its government textbook production and distribution requirements for the peak period.According to the company's internal performance evaluation, this represented the first time since the company's establishment in 1997 that all peak-season obligations had been completed ahead of schedule.The internal evaluation estimated:approximately 18–20% overall operational improvement, associated with increased productivity and production efficiency;approximately 15–20% reduction in equipment-related stoppages and failures during the peak-production period.The structured methodology introduced during the period continued to be used within the organization in subsequent years.These results should not be interpreted as evidence that vibration monitoring, operator feedback, or any other single technique independently generated the full improvement.The relevant observation was that several relatively inexpensive practices became more effective when they were linked into one decision-making process.6. Discussion6.1 Condition Monitoring Does Not Always Require a Complex SystemModern predictive-maintenance programs can include permanently installed sensors, cloud platforms, artificial intelligence, sophisticated signal processing, and large historical datasets.Those technologies can provide significant value.However, many industrial facilities already contain useful reliability information that is not being systematically connected.Examples include:operator observations;maintenance history;temperature changes;vibration changes;inspection results;repeated minor stops;equipment loading;production history.The challenge is therefore not always a lack of data.Sometimes the challenge is that useful information exists in separate places and does not reach the people making maintenance decisions.Condition-based maintenance research similarly treats data acquisition, processing, diagnostics, and maintenance decisions as connected stages rather than independent activities [1].6.2 Baseline Behavior MattersA measurement becomes much more useful when there is a meaningful reference point.Knowing that a machine has a particular vibration or temperature value may provide limited information by itself.Knowing that the same machine is now behaving differently from its normal condition may provide a reason for further investigation.This is particularly important in practical industrial maintenance, where nominally similar machines may behave differently due to:installation conditions;age;loading;alignment;operating speed;maintenance history.The objective is therefore not only to define universal alarm values.It is also to understand normal behavior for the equipment being monitored.6.3 Small Events Can Contain Useful InformationLarge breakdowns naturally receive attention.Small interruptions often do not.But a major failure is often the end of a deterioration process rather than its beginning.A five-minute interruption may not matter.Ten similar interruptions affecting the same machine over several shifts may matter considerably.This is why small stoppages should be examined for frequency, repetition, and pattern, rather than evaluated only by their individual duration.The broader maintenance-management literature likewise emphasizes the importance of information systems, maintenance techniques, planning, and scheduling in supporting effective maintenance decisions [7].6.4 Operators Are Part of the Reliability SystemMaintenance technicians cannot continuously observe every production machine.Operators effectively do.That makes operator observations valuable, provided there is a process for collecting and evaluating them.This does not mean replacing technical inspection with subjective impressions.It means adding operator observations to other evidence such as:maintenance history;inspections;vibration;temperature;recurring stoppages.The philosophy is compatible with TPM approaches, which emphasize organizational participation in equipment effectiveness and maintenance rather than separating production and maintenance completely [4].6.5 Maintenance Performance and Manufacturing Performance Are ConnectedOne reason maintenance sometimes receives insufficient attention is that it may be treated primarily as a cost center.In practice, equipment reliability affects:production volume;product quality;delivery performance;overtime;spare-parts consumption;customer commitments.Research by McKone, Schroeder, and Cua found a significant relationship between Total Productive Maintenance practices and several dimensions of manufacturing performance, including quality and delivery performance [5].That relationship was also visible in our case.The purpose of the maintenance changes was not simply to reduce the number of repairs.The larger objective was to protect the production schedule.7. A Practical Framework for High-Load ManufacturingBased on this experience, I would suggest seven questions for maintenance and production teams operating under sustained production loads.1. What does normal equipment behavior look like?Without a baseline, abnormal behavior becomes more difficult to identify.2. Are operator observations reaching maintenance personnel?Operators may detect subtle changes before scheduled inspections do.3. Are minor stoppages being recorded?Recurring small events may reveal deterioration before a major failure occurs.4. Are changes in vibration and temperature being investigated?Condition changes can provide useful early-warning information.5. Does the maintenance plan reflect actual equipment utilization?A calendar does not always reflect real mechanical loading.6. Can production sequencing respond to equipment condition?A temporary production adjustment may create enough time to reach a planned maintenance window.7. Can the team intervene before production is forced to stop?That should ultimately be the objective.8. Limitations of This Case StudyThis case should be understood as a real-world industrial operational case study, not as a controlled scientific experiment.The performance figures reported here were based on the company's internal operational evaluation, production information, and maintenance experience during the period.The work did not use:randomized control groups;controlled laboratory testing;identical parallel production systems;experimental isolation of individual maintenance variables.Several factors can affect productivity and equipment reliability simultaneously.For that reason, the reported results should not be interpreted as demonstrating that any single maintenance technique independently caused the entire 18–20% operational improvement or 15–20% reduction in equipment-related stoppages and failures.The case instead documents the operational results observed after a coordinated maintenance and production-management approach was implemented during a real high-load manufacturing period.This limitation is important because industrial facilities rarely operate under controlled experimental conditions.Real manufacturing decisions are made while production is running, customer deadlines remain active, machinery continues aging, and multiple operational variables change simultaneously.The value of an industrial case study is therefore different from that of a controlled experiment.Its purpose is to describe the problem, intervention, observed results, and lessons that may be tested or adapted in other operating environments.9. Broader ApplicabilityAlthough this case involved industrial printing equipment, the underlying principles are not specific to the printing industry.Similar reliability challenges occur in:automated packaging;food and beverage production;material handling;high-speed manufacturing;assembly equipment;robotics;other mechanically intensive production environments.The exact monitoring parameters and maintenance practices will differ between industries.However, the general process remains applicable:establish normal equipment behavior;capture observations from people closest to the machinery;record recurring abnormalities;monitor meaningful physical condition indicators;evaluate equipment condition in relation to production demand;create an intervention window before failure whenever possible.Condition-based maintenance literature provides a formal engineering framework for many of these principles [1,2], while research on integrated production and maintenance planning demonstrates why the relationship between the two functions deserves explicit consideration [6].10. ConclusionIndustrial reliability under extreme production load requires a different mindset from conventional reactive maintenance.When equipment operates almost continuously, waiting for a machine to fail can be too expensive.The approach used during this project did not depend on one sophisticated technology.Instead, it integrated several practical sources of information:operator feedback;structured inspection;vibration monitoring;temperature monitoring;minor-stoppage recording;preventive maintenance;production planning.Together, these practices provided a clearer view of changing equipment behavior and created opportunities to intervene before production was forced to stop.The broader lesson I took from this experience is simple:Reliability is not only a maintenance function. It is an operating strategy.Maintenance, production planning, and machine operation should not function as isolated systems.They are different sources of information about the same production process.When that information is combined effectively, maintenance can move away from asking:“How quickly can we repair the machine after it stops?”and toward the more useful question:“What is the machine telling us before it stops?”References[1] Jardine, A.K.S., Lin, D., & Banjevic, D. (2006). A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mechanical Systems and Signal Processing, 20(7), 1483–1510.[2] International Organization for Standardization. (2018). ISO 17359:2018 — Condition monitoring and diagnostics of machines — General guidelines. Third Edition. ISO. The standard was reviewed and confirmed in 2023 and remains current.[3] Randall, R.B., & Antoni, J. (2011). Rolling element bearing diagnostics—A tutorial. Mechanical Systems and Signal Processing, 25(2), 485–520.[4] Ahuja, I.P.S., & Khamba, J.S. (2008). Total productive maintenance: literature review and directions. International Journal of Quality & Reliability Management, 25(7), 709–756.[5] McKone, K.E., Schroeder, R.G., & Cua, K.O. (2001). The impact of total productive maintenance practices on manufacturing performance. Journal of Operations Management, 19(1), 39–58.[6] Nourelfath, M., Fitouhi, M.-C., & Machani, M. (2010). An Integrated Model for Production and Preventive Maintenance Planning in Multi-State Systems. IEEE Transactions on Reliability, 59(3), 496–506.[7] Garg, A., & Deshmukh, S.G. (2006). Maintenance management: literature review and directions. Journal of Quality in Maintenance Engineering, 12(3), 205–238.