ServicesStage 04 · Operate07 / 11

Edge Intelligence, Control and Automation

Bring analysis closer to the equipment and decisions it supports

Closed-loop step responsepressure control · PI retuned
010 s5 s10 s15 s20 s±2% bandrise time (90%) · 1.9 speak · 1.234 · overshoot 23.4%settling (±2%) · 7.7 st 0 s · y 0 · setpoint 0t 1.0 s · y 0 · setpoint 0t 2.1 s · y 0.011 · setpoint 1t 3.1 s · y 0.507 · setpoint 1t 4.2 s · y 1.058 · setpoint 1t 5.3 s · y 1.231 · setpoint 1t 6.3 s · y 1.115 · setpoint 1t 7.4 s · y 0.976 · setpoint 1t 8.4 s · y 0.946 · setpoint 1t 9.4 s · y 0.973 · setpoint 1t 10.6 s · y 1.005 · setpoint 1t 11.6 s · y 1.013 · setpoint 1t 12.6 s · y 1.006 · setpoint 1t 13.7 s · y 0.999 · setpoint 1t 14.7 s · y 0.997 · setpoint 1t 15.8 s · y 0.999 · setpoint 1t 16.9 s · y 1.000 · setpoint 1t 17.9 s · y 1.001 · setpoint 1t 19.0 s · y 1.000 · setpoint 1t 20 s · y 1.000 · setpoint 1
normalised outputovershoot 23.4% · settling 7.7 s

Edge intelligence and control

MNC Analytics develops systems that process information locally and support timely responses to changing conditions. We deploy analytics on embedded processors, industrial computers and edge gateways where latency, connectivity, bandwidth or operational requirements favour local computation.

We also develop control and automation solutions that translate system models and measurements into defined actions. The required level of autonomy is established around the operating context and the responsibilities of the people using the system.

What we offer

8

  1. 01

    Deployment and optimisation of analytical models on edge devices

  2. 02

    Assessment of accuracy, response time, power and memory requirements

  3. 03

    Offline operation, buffering, diagnostics and recovery logic

  4. 04

    Dynamic modelling and identification of controlled systems

  5. 05

    Controller design and tuning, including conventional, state-space and model predictive approaches

  6. 06

    Integration with programmable logic controllers, supervisory systems and operator interfaces

  7. 07

    Software-in-the-loop and hardware-in-the-loop testing

  8. 08

    Defined overrides, interlocks, fallback behaviour and release procedures

What this enables

Local monitoring, faster decision support and controlled automation aligned with equipment and operating constraints. We define whether outputs inform an operator, recommend an action or form part of an authorised control function.

Related work

All work
01Manufacturing and industrial operationsGermany

Automated visual inspection of extruded aluminium profiles

A line-scan camera station with controlled lighting and an edge inference unit now grades surface defects at full line speed, replacing sampled manual inspection.

  • 100%of production inspected, at line speed
  • -84%customer-reported surface defects
  • < 40 msinference per frame on the edge unit
Read more
Customer-reported surface defects per monthClaims
-100102030M1M2M3M4M5M6M7M8M9M10M1 Claims: 19 claimsM2 Claims: 17 claimsM3 Claims: 21 claimsM4 Claims: 16 claimsM5 Claims: 14 claimsM6 Claims: 9 claimsM7 Claims: 6 claimsM8 Claims: 4 claimsM9 Claims: 3 claimsM10 Claims: 3 claims
claimsM10 · 3 claims
02Agriculture, water and environmental systemsSaudi Arabia

Optimising pump scheduling across a water transmission network

A hydraulic model and a mixed-integer optimiser schedule pumps against tariff periods and reservoir constraints, cutting energy cost while keeping storage within agreed limits.

  • -16%pumping energy cost
  • -38%pump starts per day
  • 0reservoir limit breaches in 9 months
Read more
Monthly pumping energy cost, indexed to baselineBaselineOptimised
020406080100M1M2M3M4M5M6M1 Baseline: 100 indexM1 Optimised: 96 indexM2 Baseline: 100 indexM2 Optimised: 90 indexM3 Baseline: 100 indexM3 Optimised: 86 indexM4 Baseline: 100 indexM4 Optimised: 84 indexM5 Baseline: 100 indexM5 Optimised: 83 indexM6 Baseline: 100 indexM6 Optimised: 84 index
indexM6 · 84 index
Next service · 08Simulation and Digital Twins

Contact

Tell us about the problem you need to solve

You may have a defined project, an underperforming process or a technical question that needs investigation. We can help assess the challenge, identify the information required and define an appropriate next step.

Discuss your projecthq@mncanalytics.com