Integrated Automation

MOM, MES and AGV, the layer that turns a collection of machines into a plant that can be measured, traced and improved.

Architecture

One architecture, five layers

We follow the ISA-95 reference model, so responsibilities are cleanly separated and integration with your existing ERP, or with a third-party MES you already run, is a defined interface rather than a custom project.

ISA-95 automation architecture: ERP, MOM, MES, SCADA, PLC, equipment and AGV fleet
Layer 01

MOM — Manufacturing Operations Management

MOM sits above MES and manages operations as a whole rather than execution on a single line. It is where multi-line and multi-site questions get answered.

Production Operations

Order orchestration across lines, capacity allocation and production planning against demand.

Quality Operations

Specification management, non-conformance workflow, corrective action tracking and audit readiness.

Maintenance Operations

Planned and condition-based maintenance, spare parts management, equipment history.

Inventory Operations

Material and WIP visibility across the plant, consumption tracking and reconciliation.

Layer 02

MES — Manufacturing Execution

MES runs the line: it dispatches work, enforces the process, captures what happened and makes the result analysable. Everything downstream in quality and cost analysis depends on the completeness of this capture.

WIP Tracking

Real-time work-in-progress position by station, lot and unit across the whole line.

Recipe & Parameter Control

Controlled process parameters downloaded to equipment, with change management and version history.

Unit-Level Genealogy

Full traceability from wafer or cell through to finished, serialised product.

Statistical Process Control

Inline metrology feeding SPC charts with automatic out-of-control alerting.

Yield & Loss Analysis

Yield by station, defect Pareto and loss attribution to specific equipment or parameters.

OEE & Performance

Overall equipment effectiveness with availability, performance and quality decomposition.

Dispatching & Scheduling

Work order dispatch with material availability and equipment state awareness.

Alarm Management

Consolidated alarm handling with escalation paths and response tracking.

Cost Attribution

Labour, material and overhead allocated by batch, product and customer for real cost visibility.

Traceability is designed in, or it is not there

Unit-level genealogy depends on identity being readable at every transfer point and on equipment being able to report what it did. Both are specification decisions made when the line is designed. Retrofitting traceability onto a line that was not built for it typically yields partial data with gaps exactly where investigations need it most.

Layer 03

AGV and AMR material transport

Automated transport replaces the trolleys, and with them the handling damage, the queueing at load ports and the gaps in the traceability record. Two vehicle classes do this job and they are not interchangeable. Choosing between them per route is part of the layout work, not a purchasing decision taken afterwards.

AMR An autonomous mobile robot scanning its surroundings, its route curving around a pallet left in the aisle and rejoining the far side

Autonomous mobile robot

Carries its own map. It localises by laser scan against the building it can see, so a route is a piece of software rather than a piece of floor. Meeting the pallet above, it plans around it and carries on, which is what makes it usable in aisles shared with people and forklifts.

  • Laser SLAM
  • Free navigation
  • Re-routes around obstacles
  • 360° safety scan
AGV A guided vehicle on a fixed floor path, halted by a pallet standing on the route it cannot leave, with a warning raised

Automated guided vehicle

Follows a path laid into the floor, as magnetic tape, inductive wire or markers. It cannot leave that path, so the pallet above stops it and raises an alarm. On a trunk route that will run unchanged for years, that constraint buys higher payload, a simpler fleet and a lower cost per vehicle.

  • Fixed guide path
  • Stops at obstacles
  • High payload
  • Lower cost per unit
Most plants end up with both. The trunk routes justify guided vehicles; the areas that keep changing justify autonomous ones.
AGV — guidedAMR — autonomous
How it navigatesFollows a fixed path: magnetic tape, inductive wire, or floor markersBuilds and localises against its own map using laser SLAM
Changing a routePhysical work. The guide path is re-laid and the floor re-commissionedA software change. The map is edited and the fleet redeployed
Meeting an obstacleStops and waits. It cannot leave the pathPlans around it and continues, if the aisle allows
Best suited toHigh-volume fixed routes that will not move for yearsMixed routes, shared aisles, and layouts still expected to change
Relative cost per vehicleLower, with cost in the floor infrastructureHigher per unit, with no floor infrastructure
Where we use itLong trunk routes between process areas, and cassette loops that mirror the process flowFeeding stations that get reconfigured, and any area shared with people and forklifts

How we deliver it on a production line

Transport is designed with the line, at the same time as the equipment layout. Retrofitting it into a plant that was laid out without it is where most of the cost and disappointment comes from.

Corridors in the Layout

Transport routes, passing places and buffer positions drawn at layout stage, so vehicles are not squeezed into whatever floor the equipment left behind.

Fleet Sizing

Vehicle count derived from takt time, route length, dwell at load ports and charging strategy, rather than estimated and corrected later.

Load-Port Docking

Docking positions and tolerances specified with the equipment vendors, so loaders and unloaders are ordered AGV-ready instead of modified on site.

Traffic & Charging

Fleet management for routing, right of way at junctions, opportunity charging and a battery strategy that holds availability across all shifts.

Safety Engineering

Certified 360° laser scanners, zone speed limits, floor markings and interlocks with equipment doors, designed to the plant's safety assessment.

MES Integration

Every move dispatched by, and reported back to, the MES layer, so material movement forms part of the genealogy rather than a blind spot in it.

The interface that is usually missed

A vehicle can reach a machine and still not be able to hand material to it. Load-port height, docking tolerance, handshake signals and the cassette or tray standard all have to be agreed across every equipment vendor on the line. Doing that during specification costs a series of conversations. Doing it after delivery costs mechanical rework on machines that are already installed and commissioned.

Layer 04

Warehousing, line control and integration

Interfaces defined during line engineering so equipment from multiple vendors integrates to one architecture.
ComponentFunctionTypical interfaces
WMS, warehouse managementInventory location, picking, replenishment and stock accuracyMOM, MES, AGV fleet manager, ERP
Intelligent logistics (ILD)Automated material distribution between storage and lineAGV fleet, stacker cranes, AS/RS
Line control / SCADASupervisory control, alarm consolidation, equipment coordinationOPC-UA, SECS/GEM, PLC networks
Equipment integrationStandardised data capture and command interface per machineOPC-UA, SECS/GEM, EAP adapters
Dashboards & reportingLive plant status and management reportingMOM, MES data layer
A SIEMER MOM and MES plant overview on a tablet, showing live capacity, pass rate, per-plant output and energy monitoring
Outcome

What the automation layer actually delivers

Automation is usually justified on labour reduction, because that is the number easiest to put in a business case. It is rarely where the return ends up. The durable value is that the plant can answer questions about itself, and answer them in hours rather than after a fortnight of arguing across departments.

Two views the automation layer produces: a Pareto chart ranking stations by their share of yield loss, and a genealogy trace following one failed module back through string, cell lot, station and process parameter to root cause

Without the layer

Yield is a monthly number. Nobody agrees which station causes it, because nobody measured per station. A customer complaint means quarantining everything shipped that week, because there is no way to identify which units share the suspect lot. Recipe changes live in an operator's memory.

With it

Yield is a Pareto by station and shift. The suspect units are a query, not a quarantine. Recipes are versioned, downloaded to equipment and enforced. Cost is attributed per batch and customer, so the margin on each product is known rather than assumed.

Consistency

Enforced recipes and parameters, so output does not depend on which shift ran it.

Containment

Unit-level genealogy turns a recall into a list of affected serial numbers.

Ramp Speed

SPC and yield attribution from day one, which shortens the climb to qualified yield.

Reduced Handling

Fewer manual transfers, and less of the breakage and contamination that comes with them.

Cost Visibility

Labour, material and overhead attributed per batch, product and customer.

Audit Evidence

Warranty and compliance records retained for the product's service life.

The number that is hard to put in a business case

Labour saving is easy to calculate and usually the smaller figure. The larger one is what a single containment event costs when you cannot identify the affected units, or what a slow ramp costs in lost production before qualified yield is reached. Both are avoided by data captured automatically at every station, which only exists if the line was designed to capture it.

Applies To

Every line we engineer

Planning an automation architecture?

Whether it is a new plant or an existing line that needs a traceability layer, we can scope the MOM, MES and AGV design with you.

Contact Us Now