One shared view of your production.
production.AI links the data of your production with its context: which machine, which order, which shift, which tool, which cause, and what it costs.
Only this context makes analyses traceable and actions justifiable. Improvement builds on it, with AI as support and people who decide.
Four areas, one context
Machines & shop floor
What is running, what is down and why, straight from the machine and the operator.
- Machine connectivity independent of the manufacturer, including mixed machine parks
- Run times, downtime and OEE per machine, shift and order
- Shop floor data collection: reports and piece counts at the workstation
- Operators record causes directly at the machine
- Changeover18 min07:10 · M. Kaya
- Tool cleaning12 min08:30 · M. Kaya
- Material shortage24 min09:42 · M. Kaya
- Coil change28 min11:30 · M. Kaya
- Ejector sensor14 min12:51 · S. Wolf
Processes & resources
Process, quality and tool data across the entire life cycle.
- Process and quality data linked to order and shift
- Tool management with tool history and tryout
- Maintenance and servicing
- Record media consumption and assign it to orders
- 29 OctFA-20418 on press 3Production1,240 strokes · scrap 3.1 % · running
- 14 OctFA-20377 on press 3Production7,400 strokes · scrap 1.9 %
- 8 OctCutting edge station 4 chippedRepairInsert replaced · 6 h downtime · T. Brandt
- 2 OctFA-20341 on press 1Production8,600 strokes · scrap 1.6 %
- 30 SeptScheduled maintenance after 20,000 strokesMaintenancePunches reground · 3.5 h · T. Brandt
- 17 SeptTryout after change to station 2Tryout40 parts · released by QA
- Part
- Side panel left
- Stations
- 6
- Location
- Rack B-04
- Presses
- Press 1, Press 3
- Since
- 03/2021
| Order | Press | Strokes | Scrap |
|---|---|---|---|
| FA-2041829 Oct | Press 3 | 1,240 | 3.1 % |
| FA-2037714 Oct | Press 3 | 7,400 | 1.9 % |
| FA-203412 Oct | Press 1 | 8,600 | 1.6 % |
| FA-2029622 Sept | Press 3 | 8,100 | 1.7 % |
| FA-202548 Sept | Press 3 | 9,300 | 2 % |
Business context
Orders, feedback and costs in the same picture.
- Orders and feedback reconciled with the ERP
- Product and project data
- Economic evaluation of losses and actions
| Metric | Plan | Actual |
|---|---|---|
| Good parts | 1,400 pcs | 1,202 pcs |
| Changeover time | 15 min | 18 min |
| Unplanned downtime | – | 78 min |
| Scrap | 1.5 % | 3.1 % |
| Material | 1.9 t | 1.9 t |
- Scrap€129
- Coil change€120
- Material shortage€103
- Changeover€77
- Sensor fault€60
- Tool cleaning€52
- S. WolfToday 12:53“Ejector sensor dirty, cleaned, running again.”WZ-17Fault
- M. KayaToday 11:58“Coil change delayed, no forklift available. New coil at the press only at 11:50.”Press 3FA-20418Coil change
- M. KayaToday 09:44“Material not in place, requested supply from store 2.”Press 3Material shortage
- T. Brandt8 Oct 14:20“Cutting edge st. 4 chipped. Twice already with coils from supplier B.”WZ-17RepairSupplier B
- Forklift / supply9
- Coil from supplier B6
- Ejector sensor4
- Setup trolley incomplete3
Improvement & AI
From insight to an action that is followed up. The AI supports, your team decides.
- IDA, the issue and action management: owners, due dates, status
- Shop floor chat: questions in natural language to your own production data
- AI analytics: patterns across downtime, quality and tool data
17 of 23 cases in the last 4 weeks fell in the early shift between 9 and 11. In 14 of them the material came from store 2.
From around 15,000 strokes after maintenance, scrap with WZ-17 rises from 1.7 % to 3.1 %. The tool currently stands at 17,240 strokes.
Around €2,940 for 11.4 hours of downtime. 61 % of that falls on press 3.
One stoppage, start to finish
- Machines & shop floor
Press 3 has been down since 14:12. The platform detects the stoppage automatically, before anyone reports it.
- Machines & shop floor
The operator selects the cause at the terminal: material shortage.
- Processes & resources
The stoppage is linked to order FA-20418 and tool WZ-17.
- Improvement & AI
The shift supervisor creates an action in IDA, with an owner and a due date.
- Business context
Downtime and loss of press 3 show whether the action is working.
Each station uses a different part of the platform. Together they turn a downtime figure into an improvement.
Frequently asked questions from IT and production
Machine controls and sensors via common industrial protocols, independent of the manufacturer and also in mixed machine parks. In addition, shop floor terminals, quality and tool systems as well as ERP systems via standard interfaces. We clarify together which protocols and systems are in use at your site.
Access is granted by role, from the operator at the workstation to the management. Each role sees what it needs for its decisions.
We need access to the relevant machines and a server in your OT network that runs the connection. Whether production.AI is then operated in the cloud or on your premises determines the further scope. We clarify that in the discovery call.
Both are possible: production.AI can be operated in the cloud or on-premise in your own infrastructure. Which model suits you depends on your IT and data retention requirements and on the use case. We define that together with you.
production.AI is certified to ISO/IEC 27001 for the development, operation and support of the software. Data is transmitted and stored encrypted. Whether it resides in the cloud or in your own infrastructure is a decision we take together with you, in line with your requirements and the use case.
Read the information security policyYou start with one line or one area. That keeps the effort manageable, and the roll-out follows once the first actions take effect.