event model

An event model is a structured description of how events are represented, related, and stored in a system so that different applications can interpret, exchange, and analyze them consistently. In industrial and manufacturing environments, it commonly refers to the agreed format and rules for recording time-based occurrences such as equipment state changes, alarms, operator actions, and production milestones.

Key characteristics

A practical event model in manufacturing and OT/IT systems typically defines:

• Event types (for example, equipment state change, alarm, production start/stop, quality decision).
• Required attributes such as timestamps, resource identifiers, product or batch identifiers, and user or system source.
• Optional attributes like reason codes, severity, duration, and links to related records (work orders, NCRs, maintenance orders).
• Cardinality and rules, for example only one active equipment state event per resource at a time, or how overlapping events are handled.
• Semantics, meaning how an event should be interpreted for KPIs, traceability, or workflows (e.g., which states count as “productive time” for OEE or ISO 22400 KPIs).

Role in manufacturing and industrial systems

In OT and MES/ERP-integrated landscapes, an event model provides a common language for:

• Equipment and automation (PLC/SCADA) to report machine states and alarms in a normalized way.
• MES and production systems to track operations, resource utilization, and shift performance based on time-stamped events.
• Quality and compliance systems to link events to lots, batches, NCRs, or deviations for investigations and evidence trails.
• Operations intelligence and KPIs such as OEE, non-productive time (NPT), and ISO 22400 metrics that depend on consistent state and timestamp logic.

Event model vs data model

An event model focuses on time-ordered changes and occurrences, while a general data model describes relatively static entities such as products, equipment, or bills of material. In many architectures, the data model defines what things exist, and the event model defines what happens to them and when.

Common confusion

• Event model vs message schema: A message schema (for example, a JSON or XML structure) defines the technical format of a single message. An event model is broader and covers event types, their relationships, sequencing rules, and interpretation for analytics and workflows.
• Event model vs process model: A process model describes the intended sequence of activities or states. An event model describes the actual recorded occurrences that can later be compared against the process model.

Relation to ISO 22400 KPIs and equipment states

When implementing ISO 22400 manufacturing KPIs, an event model is often used to standardize how equipment state events are captured. This includes defining a unified set of states, ensuring only one active state per resource at a time, capturing precise start and end timestamps, and recording reason codes for state changes. Such a model allows vendor-specific SCADA or MES states to be normalized so that KPIs are calculated consistently across lines, plants, and systems.