Offer · Maintenance on the shop floor from a phone: fewer breakdowns, planned work instead of firefighting
KaizoCMMS
A CMMS for the maintenance team of a production plant: machine registry with a QR code on every machine, breakdown reports and work orders from a phone, planned inspections, operator checklists, spare parts and failure analysis. Local AI suggests likely causes and when to plan the next inspection. Everything runs in your plant network.
For whom
Maintenance managers, technicians (mechanics, electricians, automation engineers) and machine operators in mid-sized production plants that today keep breakdowns and inspections in Excel, on paper or in a notebook.
Runs on
AI Office Box Start or Pro in the plant network; technicians use their phones (PWA), operators a tablet at the machine. Compare AI Box packages →
System diagram
Inputs, outputs and interfaces
What goes into the AI Office Box in your plant, what comes out, and through which interface. Machine data, work orders and costs stay in your network.
Your network: your data stays here
Inputs
Breakdown reports from the floor
An operator or technician scans the QR code on the machine and reports the problem with a photo or a voice note.
Phone or tablet (PWA), also offlineMachine registry and documentation
Plant → line → machine → assembly, with manuals, criticality and history. Imported from your Excel lists.
Excel / CSV import, web appOperating hours and counters
Hours, cycles and alarms that trigger planned inspections; optional at the start.
Manual entry, PLC or SCADA: OPC UA, Modbus
AI Office Box Start / Pro in the plant
- 1Work order flow: report, assign, carry out, approve; time and parts used
- 2Planned maintenance schedules by calendar, operating hours or counters
- 3Jishu Hozen: operator CILT checklists and defect tags
- 4Spare parts store: stock, minimum levels, reservations per order
- 5Failure analysis: MTBF, MTTR, availability, repeat failures
- 6Local language model: likely causes from the history and when to plan an inspection
Outputs
Work orders for technicians
The order on the technician's phone with the machine history; closed on the spot, also without signal on the hall.
Phone app (PWA), push notificationsInspection plan and checklists
Automatic orders for planned inspections and daily CILT checklists for operators.
Web app, tablet at the machineSpare parts orders
Parts below the minimum level listed for purchase, with the orders that used them.
Report, e-mail or ERP exportKPI dashboard for the maintenance manager
Backlog, planned vs reactive work, MTBF, MTTR, availability and cost per machine.
Web dashboard in your LAN
Learning loop: every closed work order (cause, repair, parts, time) feeds the failure history, so the cause suggestions and inspection intervals get better every month.
Interfaces
| Interface | Direction | Protocol / format | What passes |
|---|---|---|---|
| Phones and tablets | Floor ↔ AI Box | PWA over your Wi-Fi, works offline and syncs later | Reports, photos, work orders, checklists |
| Machine registry | Your files → AI Box | Excel / CSV import, then the web app | Machines, assemblies, documentation, criticality |
| Counters and alarms | PLC / SCADA → AI Box | OPC UA or Modbus, optional, agreed with your automation team | Operating hours, cycles, alarms |
| ERP | AI Box ↔ ERP | File export or API, depending on your ERP | Spare parts, purchase requests, maintenance costs |
| KPI dashboard | AI Box → maintenance manager | HTTPS in your LAN, any browser | Backlog, MTBF, MTTR, availability, costs |
| Remote service | Our team ↔ AI Box | VPN, under the SLA | Updates, backups, support |
What it does
Eight modules, one app for the whole maintenance team
Machine registry
Plant, line, machine and assembly, a QR code on every machine, documentation and criticality.
Work orders
Report with a photo, assign, carry out, approve. Time and parts are recorded on the order.
Planned maintenance
Inspections by calendar, operating hours or counters; the orders are created automatically.
Jishu Hozen
Operator CILT checklists (clean, inspect, lubricate, tighten) and defect tags with photos.
Technician app
Scan the QR code, see the machine history, close the order, also without signal on the hall.
Spare parts
Stock, minimum levels, reservations per work order and purchase lists.
Failure analysis
MTBF, MTTR, availability and repeat failures; AI suggests likely causes and inspection timing.
Reports and KPIs
Backlog, planned vs reactive work and cost per machine for the maintenance manager.
What we measure
KPIs we measure before the pilot and prove after it
We measure these with you before the pilot (the baseline), so the return can be proven in złoty afterwards.
| KPI | Definition | How it turns into money |
|---|---|---|
| Unplanned downtime | Hours a machine stops because of a breakdown | Lost output per hour × hours; the largest item in most plants |
| MTBF | Mean time between failures | Fewer breakdowns, fewer emergency call-outs and overtime |
| MTTR | Mean time to repair | The technician finds the machine history and parts faster |
| Planned work share | % of maintenance hours that were planned | Planned work is cheaper and does not stop production unexpectedly |
| Backlog | Open work orders and their age | Shows whether the team keeps up and where it needs help |
| Spare parts cost | Parts stock value and emergency purchases | Less frozen stock, fewer rush orders |
| OEE availability | Share of planned time the machine runs | The overall effect, for the plant director |
Rollout
Five stages, each with a result you can check
- 11–2 weeks
Discovery
Interview with the maintenance manager, choice of one hall or line, baseline KPIs from your records. Result: agreed KPIs and success criteria.
- 21–2 weeks
Machine registry
Import of your machine lists, QR codes on the machines, critical machines marked. Result: the registry in the app.
- 34–6 weeks
Pilot
Breakdown reports, work orders and inspections for the pilot area, on the technicians' phones. Result: the first real MTBF, MTTR and backlog.
- 44–8 weeks
Whole plant
All halls, operator checklists, spare parts store and optional PLC counters. Result: one system for the whole maintenance team.
- 5from month 3
AI and predictive maintenance
With enough history, the local model suggests causes and inspection intervals. Result: more planned work, fewer breakdowns.
Return on investment
Example calculation
Illustrative assumptions; in the discovery call we replace them with your baseline numbers.
| Item | Assumption | Saving per year |
|---|---|---|
| Less unplanned downtime | 200 hours a year × PLN 1,500 per hour, reduced by 25% | PLN 75,000 |
| Shorter repairs | 1,500 repair hours a year × PLN 120, 15% faster | PLN 27,000 |
| Less paperwork | Maintenance manager and technicians: 10 hours a week × PLN 100 | PLN 50,000 |
| Lower spare parts cost | Fewer rush orders and less frozen stock | PLN 20,000 |
| Total benefit | PLN 172,000 | |
| Solution cost (year 1) | Pilot, rollout and subscription | about PLN 120,000–170,000 |
ROI = (downtime saved + repair time saved + admin time saved + parts saved − solution cost) / solution cost. From year 2 only the subscription remains.
All our offers run on the same on-premise platform: an AI Box in your local network with your own models. Your data does not leave the company, which matters for GDPR, trade secrets and medical data. We take care of the hardware, updates and secure remote service (VPN, SLA).
Start with KaizoCMMS
Book a short discovery call. We measure your baseline KPIs first, so the result can be proven in numbers.
Request MVP