Move beyond "fix it when it breaks" to "maintain on schedule and predict by condition." Daily inspections, weekly servicing, monthly accuracy calibration and annual overhauls are all scheduled into the system under a TPM framework. Once vibration, temperature and current sensors are connected, AI models predict remaining useful life and raise alerts before failures occur.
Covering the full chain of inspection, servicing, repair, spare parts and predictive maintenance
Operators scan a code on a PDA before starting each shift, tick off checklist items, attach photos, and automatically call for maintenance when anomalies are found.
Four types of autonomous tasks — cleaning, lubrication, tightening and adjustment — are dispatched to operators on a cycle, so operating equals maintaining.
Schedule preventive maintenance by three trigger types — time, running hours or cycle count — and auto-dispatch work to technicians when due.
A closed loop from fault report → emergency dispatch → spare-parts requisition → repair record → acceptance, with average MTTR calculated automatically.
Safety-stock alerts for critical parts, requisition records and lifecycle management; spare parts linked to the equipment BOM for easy lookup.
Feed vibration, temperature and current sensor data into AI models to predict remaining useful life and warn before a failure ever happens.
TPM assigns maintenance responsibility to operators, so "the people who use the equipment also understand it." Under the TPM framework, the system splits maintenance tasks into operator autonomous maintenance and professional maintenance — the former performed daily or per shift by operators, the latter weekly or monthly by maintenance technicians.
Bearing temp trend last 6 periods (℃)
Traditional planned maintenance replaces parts when a time or cycle threshold is reached, which may mean "replacing something that was still good" or "failing before the schedule." SBK MES connects vibration, temperature, current and oil-condition sensors, uses AI models to predict true remaining life, and makes decisions based on condition.
MTTR 0:48 · Fault code F-04 heater band
Fault occurs → operator reports via PDA → repair supervisor dispatches → technician arrives → requisitions spare parts → completes repair → operator test-runs and accepts → work order closed. Every step is timestamped, and MTBF / MTTR are calculated automatically.
Spindle vibration and temperature are connected; the RUL model predicts a bearing replacement after 200 machine hours, so parts are ordered in advance.
Operators scan-inspect injection barrel, mold temperature and oil pressure each shift, and report repairs immediately on anomalies.
Oil sensors feed moisture and metal-particle data; when thresholds are exceeded the system recommends an oil change, preventing bearing seizure.
Each feeder is tagged to record usage counts; before end of life, parts are stocked and calibration is scheduled.
Maintenance data drives OEE improvement and fixed-asset management
Sign up to try the full TPM and PdM workflow for free, or book a consultant demo