Manufacturers integrate advanced planning and scheduling (APS) software with ERP and MRP in four main ways:
- A native connector, built by the APS vendor for a specific ERP.
- An API integration, where each system calls the other’s interfaces.
- Middleware or an integration platform, which maps, transforms and moves data between systems.
- Scheduled file exchange, such as flat files, database extracts or EDI.
The right approach depends on how often data changes, how many systems exchange information, and how much control you need over data mapping and error handling. Native connectors simplify standard setups, APIs suit frequent changes, middleware keeps integration logic in one place, and scheduled files suit stable data.
ERP and MRP decide which orders and materials are needed, and production planning may check capacity at a rough-cut level. APS then sequences production against finite capacity, accounting for the constraints of individual machines and lines. Integration is how those two decisions stay in step. Larger estates often combine methods, for example an API for order changes and a nightly file for master data.
What ERP and MRP plan, and what APS adds
MRP explodes demand through the bill of materials to produce planned orders and purchase requirements, using planning lead times and material availability. Depending on the ERP system and its configuration, production planning may also include rough-cut capacity checks.
APS takes those orders and schedules them on specific resources. It works against real machine calendars, labor and skills, tooling, changeovers and material arrival dates. The result is a sequence for each resource and start and finish dates the plant can meet.
The international standard for this boundary is ISA-95, the enterprise-control integration standard. It places business planning and logistics, including ERP, at Level 4, and manufacturing operations management at Level 3. APS connects enterprise planning with detailed production scheduling and execution, so the integration design decides how well the schedule reflects both business priorities and shop-floor conditions.
Integration methods compared
| Method | How it works | Suits | Check before choosing |
| Native connector | A prebuilt link between one APS and one ERP, with fields already mapped | Plants that want the fastest start on a widely used ERP | Which ERP releases and modules it covers, and who updates it when the ERP is upgraded |
| API | Each system calls the other’s interfaces as changes happen | Data that changes often, such as order changes and confirmations | Who owns the code, and how failed calls are logged and retried |
| Middleware or integration platform | A central layer that maps fields, transforms data and routes it between systems | Estates with several ERPs, plants or execution systems | The extra platform to license and maintain |
| Scheduled file exchange | Flat files, database extracts or EDI, sent at set times | Stable master data and simpler estates | The schedule is only as current as the last file |
Event-driven integration can work alongside API and middleware approaches. Instead of relying only on scheduled checks for updates, a change such as a released order or a posted goods receipt creates an event that triggers an update to the APS. How fast that update arrives depends on the messaging setup and its rules for handling failures.
What data moves each way
| Data | Direction | Common timing | System of record |
| Item master, bills of materials, routings, work centers | ERP to APS | On change, or daily | ERP (or PLM for engineering BOMs) |
| Sales orders and forecasts | ERP to APS | On change, or hourly | ERP |
| Planned and released work orders | ERP to APS | On change, or hourly | ERP |
| Inventory, open purchase orders and expected receipts | ERP to APS | Hourly, or each shift | ERP or WMS |
| Scheduled start and finish dates | APS to ERP | Each time the schedule is published | APS for dates, ERP for the order |
| Planned order changes, such as splits and date moves | APS to ERP | After a planner approves them | ERP, once accepted |
| Calculated promise dates | APS to ERP or order management | On each enquiry | APS calculates the date; ERP or order management records the accepted commitment |
| Work in progress, completions and downtime | MES to APS | As they happen | MES |
Decide who owns each record before choosing a method
The method is the easier choice. The harder one is agreeing ownership, and three rules settle most of it:
- One owner for each data element. ERP owns the order, its quantity and its customer. APS owns the sequence and the scheduled dates. Write the split down and share it with ERP and planning teams.
- A clear write-back rule. Decide exactly what APS may change in ERP: dates only, or quantities and order splits as well. Then decide who approves each kind of change.
- A conflict rule. An order can change in ERP while the APS is re-sequencing it. Decide which change wins, and how the planner is told.
These rules carry over whatever method you choose, and they make a later change of method simpler.
Where MES fits
A schedule built from ERP data alone starts going out of date as soon as it is released. The manufacturing execution system (MES) records work in progress, downtime and completions, and the APS needs that data to replan from what happened on the line. There are three common patterns:
- MES to APS directly. Shop-floor events reach the schedule fastest.
- MES to ERP to APS. It’s simpler to manage, but each hand-off adds delay.
- APS and MES on a shared data model. This reduces the need for a separate integration layer and keeps scheduling and execution data aligned.
As an example of the third pattern: Eyelit APS is advanced planning and scheduling software that runs on the same data model as Eyelit MES, so real-time WIP, downtime, and shop-floor changes feed updated schedules back to execution. Eyelit APS replans with memory: when demand or supply changes, it minimizes disruption to the existing schedule, with the degree of allowed change configurable at any level or attribute.
Promise dates are a write-back worth planning early, because they reach customers directly. Eyelit APS delivers Available-to-Promise and Capable-to-Promise in both batch and real-time modes, with the real-time module connecting to the order management system.
For a comparison of how ten APS platforms connect planning with execution, see this comparison of the 10 best advanced planning and scheduling software platforms.
Questions to ask when choosing an approach
- Which system owns each data element, and is that written down?
- How current does each data set need to be for the schedule to hold through a shift?
- What may APS write back to ERP, and who approves each kind of change?
- Which ERP release and modules does a connector cover, and who maintains it through upgrades?
- How are failed transfers and partial loads detected, retried and reported?
- Does shop-floor data reach the schedule directly from MES, or through ERP?
- How do promise dates reach the people who quote customers?
Celeste Lefebvre
Celeste Lefebvre is Product Manager at Eyelit Technologies, where she owns the product lifecycle for Eyelit APS, the company's advanced planning and scheduling software. She began as a senior application engineer, implementing scheduling systems for manufacturers and training the planners who use them.
