Real-Time Optimization — RTO, DRTO & Economic MPC
Close the gap between feasible operation and economically optimal operation — in real time, every 30 minutes, continuously.
Book a 30-min technical fit callPhD APC/MPC · CERN · IAV · SkySails
Real-Time Optimization (RTO)
Real-Time Optimization (RTO) solves an economic optimization problem — maximize throughput or margin, minimize energy cost — on a rigorous steady-state process model, typically every 30 minutes to a few hours. The result is an optimal target for the APC/MPC layer below it.
Without RTO, operators set targets by hand, usually conservatively. With RTO, the plant keeps moving toward the economic optimum as prices, feed quality and demand change.
Problems RTO Solves
Conservative manual setpoints leave margin unused
Plants run by hand drift to setpoints that feel safe, below the true optimum for throughput or energy efficiency. RTO removes this conservatism systematically.
The optimal setpoint keeps moving
Feed composition, utility prices, ambient temperature and downstream demand change continuously. Operators cannot track these interactions by hand; RTO recalculates the optimal targets on every cycle.
APC holds setpoints but does not choose them economically
MPC/APC tracks setpoints and rejects disturbances well, but the setpoint itself may not be economically optimal. RTO supplies the economically optimal targets for APC to execute.
Steady-state models drift
A model calibrated at commissioning drifts as equipment ages, catalysts deactivate and heat exchangers foul. RTO includes parameter estimation and data reconciliation to keep the model calibrated.
Classic RTO vs. DRTO vs. Economic MPC
| Aspect | Classic RTO | DRTO / Economic MPC |
|---|---|---|
| Model type | Steady-state (LP/NLP) | Dynamic model (DAE/ODE) |
| Typical cycle time | 30 min – a few hours | Minutes |
| Transients | Waits for steady state | Optimizes through transients |
| Economic objective | Separate layer | Inside the MPC horizon |
| Computational load | Moderate | High (NLP/QP at each step) |
| Best for | Continuous, steady processes | Batch, grade changes, fast dynamics |
Industrial Applications
Refinery crude unit / crude blending
Crude selection and cut-point temperatures chosen every shift to maximize valuable products and reduce energy per barrel.
Ethylene cracker severity
Furnace severity (coil outlet temperature versus feedstock) is the main lever for ethylene yield versus by-products. RTO tracks the best severity as feedstock prices and product demand change.
Ammonia synthesis loop
Loop pressure, purge rate and recycle ratio have a multi-dimensional optimum that moves with gas price, power cost and demand. RTO updates the operating point continuously.
Buildings / district energy
Pre-cool thermal mass in low-price hours and dispatch storage for peak periods. In the OptiControl-II project (ETH Zurich), MPC ran a fully occupied Swiss office building for seven months; simulations showed significantly better energy and comfort performance than the previously installed control.
Sturzenegger et al., IEEE TCST 2016 →How We Deliver RTO
Rigorous steady-state model
A first-principles model calibrated to plant data — heat-exchanger duty, reactor conversion, separation — the foundation of every RTO layer.
Parameter estimation and reconciliation
Data reconciliation closes mass and energy balances; parameter estimation keeps the model accurate as equipment ages and feed changes.
Optimizer integration
The optimization problem (LP or NLP) runs on a dedicated compute layer, reads live historian data and writes optimal setpoints to the DCS/APC layer.
APC as setpoint executor
The RTO targets go to MPC/APC controllers that track them with constraint handling and disturbance rejection. RTO and APC form a tightly coupled two-layer system.
Important: RTO Requires a Working APC Layer
RTO needs a working APC/MPC layer below it. Without APC the plant cannot hold the setpoints RTO computes, and operators drift back to manual targets. If the plant has no APC yet, the sequence is: APC first, then RTO on top.
Learn more: Economic MPC Optimization · Industrial Process NMPC / APC
Benefits
Margin where prices and feed move
The value scales with unit throughput and with how often prices, feed and constraints change. It is estimated for your unit in the diagnostic, before any investment.
Economically optimal around the clock
No shift team can re-optimize a process with many interacting variables every time prices change. RTO does it automatically on every cycle.
Software, not new instruments
RTO runs on a standard industrial PC or existing compute infrastructure, reads from the historian and writes setpoints to APC.
Value during transients too (DRTO)
Dynamic RTO optimizes through grade changes, feed transitions and start-ups instead of waiting for steady state.
Relevant Design Patterns
Book a 30-min technical fit call
If you want to close the gap between your current operation and the economic optimum — a 30-minute call is enough to assess whether RTO is feasible and what the payback looks like.
Book a 30-min technical fit call →