Forty-two industrial-buyer terms, defined plainly.
OEE, OPC UA, HMAC, SOC 2: the shorthand the shop floor and the boardroom use, with a line back to where each one applies in Haltless.
PART 01
Overall Equipment Effectiveness
The composite manufacturing metric defined as Availability × Performance × Quality. A single 0-100% figure that distils how productively a machine runs compared to its theoretical maximum, used as the standard yardstick for line and plant-level operational excellence.
RelatedShare of planned production time the machine is running
Run Time divided by Planned Production Time. Planned stoppages such as breaks and changeovers stay out of the denominator, so OEE is not penalised for work that was always going to happen.
RelatedActual cycle speed against the ideal
The ratio of actual cycle rate to the ideal cycle rate you set per machine, equivalently (ideal cycle time × total count) divided by run time. Captures slow-cycle losses that downtime tracking alone misses, the silent five percentage points that disappear before anyone notices.
RelatedGood count over total count
The share of total output that meets specification on the first pass. Scrap and rework are tracked under reason codes that line up with the downtime taxonomy, so quality losses feed the same loss-category bucket.
RelatedMean Time Between Failures
The average operating time between two unplanned stoppages on the same asset. A higher MTBF means a more reliable machine; the metric is most informative when paired with MTTR and reason-code analysis.
RelatedMean Time To Repair
The average duration of an unplanned stoppage from the moment a fault is detected until the machine is back in production. Together with MTBF it summarises the asset's reliability and the maintenance team's responsiveness.
RelatedTotal Effective Equipment Performance
OEE multiplied by the share of calendar time that is scheduled for production. Where OEE asks how well the planned time was used, TEEP asks how much of the calendar is being used at all.
RelatedThe OEE-loss decomposition from TPM
A TPM-derived categorisation that breaks OEE losses into Breakdowns, Setup and Adjustment, Idling and Minor Stops, Reduced Speed, Quality Defects, and Reduced Yield at Startup. Codified by Seiichi Nakajima at JIPM in the 1970s and 1980s; widely referenced in ISO 22400 and the broader lean manufacturing literature.
RelatedPART 02
Open Platform Communications Unified Architecture
The IEC 62541 standard for industrial machine-to-machine communication. The default protocol the Haltless edge agent reads from, since it is vendor-neutral, secure with TLS and certificates, and ships natively on most modern PLCs.
RelatedEthernet variant of the Modicon Modbus protocol
A widely deployed Ethernet variant of the Modbus protocol, originally introduced by Modicon in 1979 and still ubiquitous in legacy PLCs. The Haltless edge agent reads holding and input registers alongside OPC UA.
RelatedOriginally MQ Telemetry Transport; now used as a standalone name
A lightweight publish-subscribe protocol favoured for IIoT and edge deployments. Often paired with Sparkplug B as a payload schema for industrial telemetry; first-class MQTT input is on the platform roadmap.
RelatedODVA industrial Ethernet over the CIP stack
An industrial Ethernet protocol managed by ODVA, built on the Common Industrial Protocol. Originally driven by Rockwell Automation and most common in Allen-Bradley and ControlLogix environments; native edge-agent coverage is on the roadmap.
RelatedNative S7-300/400/1200/1500 protocol
The native protocol used by Siemens S7-300, S7-400, S7-1200, and S7-1500 PLC families. Reachable today via OPC UA gateways and on the roadmap as a native reader on the edge agent.
RelatedPART 03
Supervisory Control and Data Acquisition
The application layer that aggregates real-time telemetry from PLCs and presents it to operators on HMIs. SCADA supervises the loop; PLCs close it. Haltless complements both by laying analytics and explainable detection on top.
RelatedProgrammable Logic Controller
The industrial computer that runs the control logic on a machine. PLCs read field-device inputs (sensors, switches) and write outputs (motors, valves) deterministically on every scan cycle.
RelatedHuman-Machine Interface
The screen on or near the machine that the operator interacts with. HMIs render the SCADA view at the asset; Haltless dashboards run on the same telemetry, one network hop away.
RelatedManufacturing Execution System
The operational layer between ERP and the shop floor that schedules production, tracks work-in-progress, and enforces quality plans. Haltless can sit alongside an MES and feed it explainable health signals without replacing it.
RelatedEnterprise Resource Planning
The system of record for finance, inventory, and procurement. Haltless ships ERP adapters for SAP, Oracle, and Microsoft Dynamics; per-release availability and feature coverage are listed in the release notes.
RelatedComputerised Maintenance Management System
The system that holds work orders, asset history, and PM schedules. Haltless ships with a built-in CMMS workflow: alerts open work orders, parts are pulled from inventory, completions are signed.
RelatedTime-series database for industrial tag data
A time-series database optimised for industrial tag data, traditionally on-premise (PI System, AVEVA Historian). Haltless can coexist with a historian as an upstream source or a downstream sink.
RelatedThe Haltless reader on your network
A small Linux process running next to your machines. It reads OPC UA, Modbus TCP, CSV, and JSON files, normalises the payloads, and forwards them to the cloud through a local SQLite buffer that survives WAN outages.
RelatedPART 04
Intervene on condition, not on the calendar
A maintenance strategy that uses condition data and analytics to predict an impending failure before any threshold is breached or downtime occurs. Where Condition-Based Maintenance acts on a threshold crossing, predictive maintenance acts on the trajectory; Haltless does this with three deterministic detectors, not opaque models.
RelatedCalendar- or usage-driven maintenance
Calendar- or usage-driven maintenance regardless of current condition. Effective for components with well-characterised wear curves; costly when applied to assets where condition data is available and predictive is feasible.
RelatedThreshold-driven on the live signal
A strategy that intervenes when a real-time condition (vibration, temperature, current draw) crosses a defined threshold. The rule-based ancestor of predictive maintenance; Haltless adds anomaly-based detection on top of the same telemetry.
RelatedFlag readings that deviate from expected behaviour
The practice of flagging readings that deviate from expected behaviour. Haltless implements three deterministic detectors (static baseline z-score, EWMA, rate-of-change) instead of opaque models, so every flag is reproducible on paper.
RelatedPART 05
Exponentially Weighted Moving Average
A moving average where recent observations are weighted more heavily than older ones. Used by Haltless to catch gradual drift in a metric (slow temperature climb, creeping cycle time) before a static threshold would fire.
RelatedThe reference behaviour each detector compares against
The reference behaviour against which current readings are compared. Haltless rebuilds a MetricBaseline per machine nightly via baseline_task.py, so a CNC at Plant A does not share thresholds with a CNC at Plant B.
RelatedPART 06
Each entry references the hash of the previous one
A sequence of records where each entry includes a cryptographic hash of the previous entry. The structure that makes the Haltless audit chain tamper-evident: mutate, delete, or reorder any row and the chain breaks at that point.
RelatedHash-based Message Authentication Code
A keyed cryptographic primitive that combines a hash function (SHA-256 in our case) with a secret key. Haltless uses HMAC-SHA256 to sign every audit-chain row, so a database compromise alone cannot forge valid entries.
RelatedTenant isolation enforced inside Postgres
A PostgreSQL feature that filters which rows a session can see or modify based on policies defined per table. Haltless enforces tenant isolation in the database layer via RLS, in addition to application-layer scoping, for defence in depth.
RelatedPART 07
AICPA Trust Services Criteria
The AICPA's framework for security, availability, processing integrity, confidentiality, and privacy controls in service organisations. Haltless aligns to the SOC 2 Trust Services Criteria; Type II observation is planned for 2026.
RelatedQuality management systems
The ISO standard for quality management systems, with §7.5.3 specifying the control of documented information. The Haltless audit chain provides the attributable, time-stamped record of operational change that ISO 9001 §7.5.3 expects.
RelatedAutomotive extension of ISO 9001
The automotive-industry extension of ISO 9001, mandatory in OEM and Tier-1 supplier plants. Inherits the ISO 9001 documented-information controls and adds product-safety and traceability requirements that the audit chain helps satisfy.
RelatedElectronic records and signatures, US life sciences
The US Food and Drug Administration regulation governing electronic records and electronic signatures in life-sciences manufacturing. §11.10(e) requires a secure, time-stamped audit trail; the Haltless chain supports customer validation efforts under this rule.
RelatedGood Practice quality guidelines
Umbrella term for the Good Practice quality guidelines in life sciences (GMP for manufacturing, GLP for labs, GCP for clinical, GDP for distribution). Each regime carries an electronic-records expectation that the audit chain helps customers satisfy.
RelatedPART 08
EU Network and Information Security Directive 2
The EU directive expanding the original NIS scope to more industries including critical manufacturing, with stricter incident-reporting and risk-management obligations. The Haltless audit chain provides the forensic evidence trail needed for NIS2 investigations.
RelatedGeneral Data Protection Regulation
The EU regulation governing personal-data processing of EU residents, in force since May 2018. Haltless minimises personal data in audit logs to user identifiers and IP addresses, and supports DPO-grade tenant-scoped exports on demand.
RelatedPersonal Information Protection Law of the PRC
China's omnibus data-protection law, effective November 2021. Customers operating in China can scope tenant data residency and processing accordingly; multi-region hosting is on the roadmap.
RelatedIndustrial data access and portability
The EU regulation on access to and use of industrial data, in force from January 2024 and applicable from September 2025. It sets out access and portability rights around IoT-generated data, which the Haltless open API and per-tenant signing keys directly support.
RelatedEU regulation on AI systems
The EU regulation on AI systems, entering full application in stages through 2026 and 2027. The Haltless deterministic detectors are out of scope of the high-risk AI obligations by design; we publish formulas, not models.
RelatedPART 09
Time-series extension to PostgreSQL
A PostgreSQL extension that turns ordinary tables into time-partitioned hypertables for high-write time-series workloads. Haltless uses TimescaleDB to store machine telemetry while keeping the rest of the schema in standard Postgres.
RelatedTimescaleDB partitioned table abstraction
A TimescaleDB construct that automatically partitions time-series data into chunks for fast ingest and pruned queries. Behaves like a regular table from SQL, scales like a partitioned one without manual partition management.
RelatedConnect Haltless to your existing PLCs, run a pilot on up to ten machines, and see the explainable health score on your own equipment. No new hardware, no proprietary sensors, no consultants.
