The data you need for production planning
Orders, routings, machines, shifts, materials: which data is genuinely needed – and why incomplete machine data need not stand in the way.
The short answer: Planning needs the data you already use for planning today: orders with quantities and dates, routings, machines and their capabilities, shift models and qualifications, and material and inventory. These usually come from ERP and MES or from Excel files. Highly detailed machine data is not required – good planning results are possible with order and master data alone.
The typical inputs
The Planning Engine processes current planning data from the systems already in use:
- ERP system: production orders, delivery dates, routings, bills of materials, inventory.
- MES: feedback and status from production, where available.
- Excel and CSV: supplementary information that is often maintained in spreadsheets today, such as qualification or set-up matrices.
At a metal-processing company, for example, production orders come from SAP and are supplemented with structured Excel data.
Master data and transaction data
It helps to distinguish between two kinds of data:
Master data changes rarely. It includes machines and their capabilities, routings with process steps and times, set-up logic, shift models and qualifications.
Transaction data changes all the time: open orders, delivery dates, current inventory, incoming material, and the availability of machines and staff.
For a first plan, a snapshot of the transaction data plus the relevant master data is often enough.
The right information in the right structure
What matters is not as much data as possible, but the right information in the right structure. Which data is actually needed follows from the planning logic agreed in the workshop: if pre-material is a start criterion, the engine needs inventory and incoming material; if qualifications decide who works at which machine, it needs the qualification matrix.
At ams OSRAM, planning builds on orders, delivery dates, pre-material, routings, machine availability, shift plans and qualifications.
Dealing with gaps and inaccuracies
Many companies hesitate because their data is not perfect. That is rarely an obstacle. Data availability and quality are reviewed together. Production scheduling does not require highly detailed machine data. Where values are missing or inaccurate, they are filled in together during the project or documented as assumptions. Parallel planning then shows whether the results hold up in practice.
Data exchange during the project
In the early project phases, data is deliberately exchanged manually or semi-automatically, for example via Excel or CSV exports. No interfaces need to be built for this yet; we clarify the effort for providing data and access together at the outset. Integration into ERP and MES only follows once you have decided to roll out. More on this in ERP and MES integration: a complement, not a replacement.
Data security
Planning only needs data that is already used for planning today. Which data is exchanged, and to what extent, is agreed together at the start of the project.
How much potential is there in your planning?
In the free potential check (45 minutes), we talk about your current planning and assess together whether and where DeepSynergy.AI Production Planning can help.
