Can you calculate the losses from a day's equipment downtime?

For a long time, equipment maintenance in Southeast Asian factories has only followed two scripts: one is "repair when it breaks down", where maintenance procedures are only initiated when the machine alarms or production stops; the other is "regular maintenance", where the machine is disassembled and oiled or parts are replaced weekly or monthly, regardless of the condition of the equipment.

Both approaches have their problems. The former is risky; if the machine stops during peak season or before the delivery date of a large order, the loss often exceeds the value of the equipment itself. The latter seems stable, but it is actually wasteful, as it allows for the replacement of parts three months in advance. However, the wear and tear that really needs attention is overlooked because the equipment has just been maintained.

Predictive maintenance (PdM) offers a third approach: using sensors to continuously monitor equipment status and automatically issue warnings when abnormal trends appear in the data, allowing maintenance teams to schedule work orders and prepare spare parts before the equipment actually fails.

The essential differences between the three maintenance modes

To determine which model is suitable for your factory, first clarify the logic behind all three:

I. Breakdown Maintenance (BM): If you don't take action until something goes wrong, the cost may seem to be the lowest, but the hidden costs of downtime losses, emergency purchase of spare parts, and overtime work are often three to five times the book repair cost.

2. Preventive Maintenance (PM): With fixed cycles and predictable execution, it is the mainstream practice in most factories in Southeast Asia. Its main drawback is "over-maintenance". Equipment is maintained according to the schedule, but it may not be necessary, resulting in the accumulation of spare parts inventory and waste of manpower.

3. Predictive Maintenance (PdM): Maintenance is determined based on actual equipment operating data, and theoretically, maintenance is only performed when "really needed". The implementation threshold is higher than the previous two, but the long-term benefits are the most significant, especially in scenarios with high equipment unit prices and large downtime losses. The payback period is usually within eighteen months.

Three starting scenarios for PdM in Southeast Asian factories

PdM (Product Manufacturing) does not require a complete factory overhaul at once. The following three scenarios are the most common starting points for factories in Southeast Asia:

1. Spindle Vibration Monitoring (CNC Machining Machines): A vibration sensor is installed on the spindle or turret to continuously record the vibration frequency and amplitude. When the value deviates from the baseline and exceeds the set threshold, the system automatically triggers an alarm, allowing maintenance personnel to schedule an inspection during the next shift break, rather than waiting until the tool breaks or the spindle burns out. This is especially useful for multi-shift metal processing plants in Vietnam.

II. Mold Temperature and Pressure Trend Monitoring (Injection Molding Machine): Unstable mold temperature and a slow drop in hydraulic pressure in the injection molding machine are often precursors to oil seal aging or scale buildup in the cooling water system. By continuously recording the trend charts of these two parameters, intervention can be initiated before the yield begins to decline, avoiding material losses caused by the scrapping of an entire batch of parts.

3. Motor Current Anomaly Detection (Conveyor Belt, Pump, Fan): Under normal load, the current should remain within a stable range. Periodic spikes or prolonged slow increases usually indicate bearing wear or loose transmission components. These sensors are low-cost and require no downtime for installation, making them the easiest starting point for small and medium-sized factories with limited resources.

PdM and MES Integration: Enabling Early Warning Systems to Truly Take Root

Sensing data itself is not equivalent to action; it requires a system that can receive, judge, and dispatch data to generate value. This is precisely the role that MES (Manufacturing Execution System) plays in the PdM architecture.

The typical integrated process is as follows: sensor data is uploaded to the IIoT platform, the anomaly detection engine compares it with a benchmark value, and the warning signal is transmitted to the device management module of the MES. The system then automatically generates a maintenance work order, assigns the responsible personnel, and locates the required spare parts. The entire process, from anomaly detection to work order generation, can be completed within minutes without relying on manual reporting.

For factories that have already implemented MES, PdM adds a layer of awareness to the existing system, rather than rebuilding the process. After the two are integrated, historical maintenance records of equipment, spare parts consumption curves, and conflicts with production scheduling can all be managed on the same interface, so that maintenance plans and production plans no longer interfere with each other.

A pragmatic path from regular maintenance.

Completely abandoning regular maintenance and fully transitioning to PdM is unrealistic, especially for Southeast Asian factories with complex equipment types and weak data foundations. A more pragmatic approach is a hybrid strategy: prioritize the implementation of PdM on high-value, high-risk equipment (such as CNC machining centers, injection molding machines, and critical conveyor lines), while maintaining the existing PM cycles for other equipment, but gradually adjusting maintenance intervals with the help of sensor data.

The recommended three steps for starting out are: First, select two to three key pieces of equipment on a production line as pilot projects, install basic sensors (vibration, temperature, current), and establish baseline data for three to six months; Second, set early warning thresholds, observe the correlation between early warnings and actual failures, and adjust the judgment logic; Third, after confirming the effectiveness of the pilot project, standardize the process and gradually expand it to other equipment.

Every step should involve the maintenance team, not just the IT or equipment engineers. A technician's intuitive assessment of the equipment's condition is often the most effective basis for adjusting sensing thresholds.

Further Reading

If you are evaluating the overall path to factory digitization, the following articles can serve as a reference:

• Three preparations that must be made before implementing MES in Southeast Asian factories

• AI-powered intelligent quality inspection x MES integrated on-site management

• OEE Metrics in Practice: The Primary Measuring Tool for Automation Effectiveness in Southeast Asian Factories