Process optimization deployments

Analysts at the KBR Innovation Studio Bengaluru will use Seeq to investigate anomalies flagged by Orbital in KBR's INSITE platform.

Borouge International ran a live proof of concept of Honeywell Experion Cognition, an AI-enabled control system platform for autonomous control room operations, at its Ruwais facility in Abu Dhabi.

Lonza developed an OEE solution in Seeq with real-time monitoring at its Portsmouth, New Hampshire facility and is expanding it globally.

Stated result asset capacity up 22 percent in the first campaign

PBF Energy uses Seeq for real-time visibility of energy efficiency across six refineries, with reusable projects on fuel firing and steam systems.

International Paper uses Seeq as part of its analytics ecosystem, supporting 15 mills, 300+ users and 48 data source connections.

Aramco uses Emerson's AI-driven Aspen Hybrid Models in its refinery planning framework, implemented for continuous catalyst regeneration and platformer units across its refining network.

Stated result Yield and quality prediction accuracy of up to 98.5% in key refinery units

Honeywell Performance+ Services, delivered through Honeywell Forge, will be deployed on core units of the Dangote refinery in Lekki, with operator training simulators built on digital twins of the facilities.

Trident Maritime Systems partnered with Gecko Robotics to deploy Gecko's robots and Cantilever software on its factory floors, across 20 fabrication facilities that supply the U.S. Navy.

PepsiCo is building digital twins of U.S. manufacturing and warehouse facilities with Siemens Digital Twin Composer to test and validate new line configurations before changing the plant.

Stated result 20% increase in throughput on initial deployment

Stellantis put Tractian energy monitoring sensors on welding robot control panels at one plant to track each robot's electrical behavior; the project is being scaled to 134 machines.

Stated result 17% reduction in energy consumption per welding point; estimated savings of 1 GWh a year.

TotalEnergies is piloting Honeywell's AI-assisted Experion Operations Assistant in the control room of the Delayed Coking Unit at its Port Arthur Refinery to forecast alarm events before they happen.

Stated result Preliminary: five potential events forecast, on average 12 minutes before an alarm incident.

Multiple coordinated autonomous control AI agents (FKDPP reinforcement learning), integrated with CENTUM VP, directly control and optimize acid gas removal operations at Aramco's Fadhili Gas Plant.

Stated result Initial results: 10% to 15% less amine and steam usage and around 5% less power usage.

Kyoto craft brewer Craft Bank and Yokogawa Digital ran a proof of concept in which FKDPP autonomous control AI generated the fermentation temperature schedule for Bank IPA; brewers applied it manually.

Stated result Fermentation time shortened from 336 hours to 240 hours, a 28% reduction.

Södra Cell ordered ABB advanced process control for the bleaching process on one production line at its Mörrum pulp mill, after earlier APC projects at its Mönsterås and Värö mills.

ClonBio signed a multi-year agreement to implement Seeq across its biorefinery assets in Europe and the U.S., starting with the Pannonia Bio grain biorefinery and ethanol facility in Hungary.

Equinor signed a multi-year agreement to implement Seeq across its global assets, initially to monitor well and process behavior.

Covestro runs Emerson's DeltaV PredictPro model predictive control on one of four polyester reactors at its Barcelona plant to manage reflux flow in the esterification batch process.

Stated result Throughput up by an estimated 1% annually; reduced reflux water use and shorter reaction duration

Tokuyama runs ABB Ability Expert Optimizer, an advanced process control system, on the calciner, kiln and cooler of its Nanyo cement plant.

Stated result Kiln thermal energy consumption down 3 percent; operator manual operations down about 70 percent

Parkland's Burnaby refinery uses Seeq for daily monitoring and incident investigation of its main air blower, sitewide inferential modeling, boilers, wet gas compressor and flare stack.

Stated result flow-anomaly analysis completed in one hour instead of more than 40 hours

ENEOS Materials officially adopted Yokogawa's FKDPP reinforcement learning AI for direct control of a distillation column at its chemical plant, after a field test of almost a year.

Stated result Steam consumption and CO2 emissions down 40% versus conventional manual control.

Votorantim Cimentos installed ABB Ability Expert Optimizer and Knowledge Manager at cement plants in Spain, Turkey, Tunisia and Morocco to optimize the process and standardize laboratory data analysis.

The Los Angeles County Sanitation Districts use AVEVA PI System, AVEVA Advanced Analytics and CONNECT, with machine learning models that optimize pumping plant operations and predict issues.

SQM centralizes nitrate plant data in AVEVA PI System and CONNECT, runs predictive and optimization models in Databricks, and shows the recommendations to operators in AVEVA PI Vision.

Stated result nitrate yield up 1%; process variability down 10-15%

TC Energy runs an AI-powered Compression Optimization Tool built on AVEVA PI System and CONNECT to optimize the compressor fleet of its natural gas pipeline network.

Stated result 180 optimization recommendations implemented between June 2023 and January 2025; 54,000 metric tons of CO2e eliminated

AtkinsRéalis deployed a digital twin of the Dragon LNG terminal in Milford Haven using AVEVA Process Simulation and AVEVA PI System to recommend process setpoints that cut energy use.

Stated result $1M energy savings identified

Lithium producer Albemarle built its AI and analytics system, branded Albemarle Intelligence, on AVEVA PI System to find process optimization opportunities across its global sites.

Stated result $150M in estimated annual improvements; 75% reduction in environmental incidents

Metallurgy group Aubert & Duval uses Braincube Real-Time Process Optimization at its Ancizes plant to adjust processes and energy use during production, and has extended it to multiple sites.

Stated result 32% improvement in process stability; 18% reduction in defects; 19% reduction in gas consumption

Saint-Gobain Weber uses Braincube Real-Time Process Optimization across 9 production sites, including on its bagging process to limit overfilling.

Stated result 90% reduction in problem-solving time for engineering teams

Pulp and paper producer Sappi uses Braincube Real-Time Process Optimization, which gives mill teams real-time operating prescriptions; 165+ optimization models are deployed across 18 mills.

Stated result 80% reduction in data preparation time

Aramco's Riyadh Refinery deployed Yokogawa's Dynamic Real-Time Optimizer (RT-OP) on its hydrocracker unit, using a high-fidelity simulation model to send optimized setpoints to advanced process control.

Barilla implemented Augury's Process Health solution (formerly Seebo), starting with one baked goods line, to find the causes of production losses from waste.

Stated result Waste reduced by 37% in a few months, per Barilla.

Lindt uses Augury's Process Health solution (formerly Seebo) on its chocolate production lines for cause analysis, throughput and quality.