D09 · PLANT-WIDE · REAL-TIME OPTIMISATION
Real-Time Process Optimisation
Continuously calculate and rank feasible operating targets as feed, demand, equipment and utility conditions change.
NOVYRON is software-first and vendor-neutral: this is configurable software, integration and lifecycle service, not process equipment. The existing PLC or DCS keeps interlocks and regulatory control, and every recommendation stays inside a site-approved operating envelope under human oversight.
THE INDUSTRIAL PROBLEM
A process can be stable and still economically suboptimal. Fixed targets cannot follow changing feed quality, energy prices, equipment availability, downstream limits or product value — and an unconstrained model will happily recommend something that is operationally impossible or unsafe.
THE NOVYRON LAYER
NOVYRON deploys a governed optimisation layer built on soft sensors, forecasts, digital twins and mathematical optimisation. It evaluates feasible setpoints against several objectives at once — quality, throughput, energy, yield and equipment stress — and returns explainable recommendations or approved supervisory targets.
TYPICAL DATA INPUTS
- Process measurements and manipulated variables
- Quality predictions and laboratory results
- Production plan, product value and demand
- Utility consumption and tariff or carbon factors
- Equipment availability and maintenance constraints
- Operator limits, recipes and safety envelopes
MODELS AND DECISIONS
- State estimation and inferential quality models
- Short-horizon demand, quality and disturbance forecasting
- Constrained nonlinear or mixed-integer optimisation
- Surrogate models where the simulation is too expensive to solve live
- Multi-objective trade-off and sensitivity analysis
- Learning from recommendation adoption and realised closed-loop performance
What each deployment reads, which models it runs, how it connects and how much authority it is given are designed for the site. None of it is pre-installed hardware, and none of it is a guaranteed autonomous function.
CORE CAPABILITIES
- EXPLICIT HARD & SOFT CONSTRAINTS
- VISIBLE OBJECTIVE WEIGHTS & ECONOMICS
- REASON, EXPECTED IMPACT & CONFIDENCE
- OPERATOR ACCEPT / REJECT FEEDBACK
- FALLBACK TO LAST APPROVED TARGETS
- REPLAY & COUNTERFACTUAL EVALUATION
HOW IT WORKS
- S1Define the objective function, the constraints and the authority boundary
- S2Develop soft sensors and forecast the disturbances that matter
- S3Calibrate the process twin, or a data-driven surrogate where it is faster
- S4Back-test recommendations against recorded operation
- S5Run shadow optimisation and review the recommendations with operators
- S6Enable bounded advisory or supervisory control, with a tested rollback
INTEGRATION POINTS
- DCS/PLC and advanced process control
- Historian, MES, scheduling and ERP
- Energy markets or utility tariff feeds
- LIMS and product genealogy
- Asset health and maintenance planning
NOVYRON connects read-only and runs in shadow mode first. Any write-back is designed, tested and approved separately. Interfaces are chosen to fit the architecture already on site rather than forcing a new hardware stack.
MEASUREMENT EXAMPLES
These are measures for a baseline and target-setting workshop, agreed from your own data. None of them is a statement of a result NOVYRON has achieved.
- Economic value per hour or per batch
- Yield and throughput at specification
- Specific energy and carbon intensity
- Constraint violation and alarm rates
- Recommendation acceptance and realised impact
- Model prediction error and optimisation solve performance
TYPICAL APPLICATIONS
- CONTINUOUS & BATCH MANUFACTURING
- UTILITIES & WATER TREATMENT
- FOOD & INGREDIENT PROCESSING
- CHEMICAL & BIOCHEMICAL PROCESSES
- MULTI-LINE PRODUCTION SCHEDULING
DEPLOYMENT PATH
- 01Objective and constraint definition
- 02Data and controllability assessment
- 03Model and soft-sensor development
- 04Historical back-test
- 05Shadow recommendations
- 06Governed advisory or supervisory rollout
WHAT STAYS OUTSIDE
Optimisation cannot override an independent protection layer. The feasible region is approved by site engineering, every model and objective version is logged, and rollback is tested before the layer is allowed to write anything.
SAFETY, QUALITY AND GOVERNANCE
- The deployment architecture can be designed to segregate customer data by tenant and deployment boundary, subject to the hosting, access-control and contractual terms agreed with you. Edge, on-premises, private-cloud and managed-cloud patterns are all evaluated against your constraints.
- Role-based access, encryption in transit and at rest, audit logging, model and version traceability and a secure update process are defined during solution design.
- Data provenance, retention and permitted model-training use are documented contractually. Customer data is not silently pooled into general models.
- Predictions carry quality flags and uncertainty where that is technically meaningful. Drift and out-of-distribution conditions trigger review rather than false precision.
Bring us the process boundary
Share the process, the signals you already collect, the control architecture, the constraints and the outcome you need. An engineer turns that into a scoped data assessment, a shadow pilot and a deployment plan.