![header image](/-/media/project/one-web/master-site/company/research-and-development/ai-solutions/header-image.jpg?h=680&iar=0&w=2560&hash=7AD7B86726921E95BC7B0FBB9DED5EFB)
Hexagon’s AI for your industry
At Hexagon, we build artificial intelligence in our technology to help our customers improve their business.
Accelerate your business with Hexagon AI
Increase efficiency
Save time and improve the accuracy of simulation, modelling and production.
Reduce cost
Unlock insights by combining data to make it accessible and useful.
Improve quality
Overcome business challenges with uniquely trained AI to suit your business needs.
Browse AI customer success stories
Ford Motor Company saves 59 days with AI engineering software
Petrochemicals leader reduces costs using physics-informed AI
AutoZone mitigates risk using AI video platform
Heavy construction uses AI reality capture to clean up drone data
Klagenfurt uses geospatial AI platform to optimise solar energy
R-evolution targets 30% more electricity efficiency with solar AI
Browse AI products by industry
Our AI products help you capture and analyse relevant datasets, model reality in a virtual environment and make the best decisions about your business. Discover how our unmatched blend of precision sensors and software can solve challenges and put AI to work for you.
ERDAS Imagine
HxGN OnCall Dispatch | Smart Advisor
LocLab
Multivista
Intergraph Smart Reference Data
ODYSSEE A-EYE
ODYSSEE CAE
ODYSSEE Platform
Nexus – Digimat
VGSTUDIO MAX
HxGN MinePlan Block Model Manager
MineProtect
Ford Motor Company saves 59 days using computer-aided engineering software with AI
“In seconds, ODYSSEE CAE provided a full 180-degree analysis that would have taken more than 1,000 hours with our previous solution.”
- Laike Misikir, Vehicle Crash Safety and CAE Engineer at Ford Motor Company
The challenge:
As engineers, Ford’s quality assurance team digitally tests new wheel designs using finite element analysis (FEA) to predict the wheel’s behaviour in different conditions and ensure the cars live up to the company’s famous slogans, “Built Ford Tough” and “Go Further.” Intensive predictive analysis of loading on the wheels can be prohibitively expensive and takes more than two months to complete a full 180-degree analysis.
The solution:
To save time and cost, Ford implemented Hexagon’s AI-enabled ODYSSEE CAE. This computer-aided engineering solution that uses AI, machine learning to reduce-order modelling for engineering simulation and production. The first step was for ODYSSEE CAE to complete 20 analyses in eight days. This process involved creating a digital simulation of the wheel to predict how it would react to forces in the real world. Real-time predictive modelling provided feedback on failure points and opportunities for optimisation before AI simulation and physical test data created alternative wheel designs.
The outcome:
With ODYSSEE CAE, Ford Motor Company’s quality assurance team can analyse eight new wheel designs in the time taken to analyse one. This method provides a huge time saving, with 92% accuracy. Hexagon’s AI solutions successfully helped Ford’s time “Go Further.”
Quicken analyses with ODYSSEE CAE
Nexus Home | Software | ODYSSEE (hexagon.com)
Petrochemicals leader reduces costs using physics-informed AI
Global petrochemicals leader SABIC used Hexagon’s AI, advanced material modelling and machine learning technologies to predict how materials responded to stress, saving time and costs.
The challenge:
It is time-consuming and costly to develop and test new ULTEM resin, an innovative thermoplastic, using classic material data generation approaches. Classical approaches are typically made up of lab testing — highly accurate but costly; advanced material modelling — reliable but complex; and AI —fast but dependent on substantial data.
The solution:
Nexus, Hexagon's digital reality platform for manufacturers, takes a more innovative, physics-informed AI approach that efficiently combines the strengths of all three classical approaches to test the plastic while minimising drawbacks such as cost, complexity and dense datasets.
A physics-based approach applies physical laws as constraints with the data fed into the model, aiming to increase accuracy. By ingesting both data and physical laws, the model can apply the laws of physics to the data in its decision-making process, instead of relying on previous results.
There were three steps to SABIC’s testing and development of their resin:
First, SABIC delivered the raw data to feed into Nexus’s AI models. The rest of the data was then generated using Digimat, Hexagon’s advanced material modelling program. This process was used to iteratively train, validate and test multiple AI models and, in the end, select the best one. Digimat’s key differentiator is its data augmentation — a statistical technique that allows the user to make estimations from incomplete data.
In the final step, the chosen AI model was trained and evaluated. This step was iterative; many models were evaluated and the best one identified based on its prediction accuracy.
The outcome:
SABIC used AI in Hexagon’s digital reality technology to efficiently combine and enrich minimal experimental data for maximum cost and time savings.
There was a significant reduction in the amount of experimental data required — 20x less than alternative lab testing and 10x less than pure AI methods.
Test and iterate faster with Nexus
Automotive retailer mitigates risk using AI video platform
“To have access to that clear information, you can’t put a number on that. That’s incredibly valuable.”
- Michael Petty, Director of Construction at AutoZone
The challenge:
AutoZone has more than 6,000 stores. Their in-house construction team needed a more effective solution to monitor activity, weather impact and safety across 170 new jobsites a year. It was prohibitively expensive and error-prone to manually monitor hundreds of sites.
The solution:
OxBlue AI video analytics platform was the answer for automatically monitoring AutoZone’s sites. Using pattern recognition, a vital component of its AI technology, OxBlue AI detected anomalies through data analysis. These analytics were transmitted to the construction team to flag patterns in progress, productivity, and safety. This accurate record of jobsite activity also provided an audit trail in case of disputes or extension requests.
The outcome:
It is now possible for AutoZone to scale a jobsite monitoring system with AI, providing a detailed, accurate and up-to-date record that can be used to make data-driven decisions and help mitigate the risk of paying for excess labour hours or multiple delays.
Gain insights with OxBlue AI
Progress Monitoring | Artificial Intelligence Cameras (oxblue.com)
Heavy construction uses AI reality capture software to organise drone data fast
![](/-/media/project/one-web/master-site/company/research-and-development/ai-solutions/heavy-construction.jpg)
“REVEAL has the intelligence and precision needed to automate the sorting and clean-up of millions of data points in just minutes.”
- Ken Fritts, Engineering Services Technical Manager at Goodfellow Bros.
The challenge:
Heavy construction contractors Goodfellow Bros. need to know what as-built conditions look like and how they affect next steps to execute projects profitably.
“As a grading contractor, we want the data to represent the dirt,” said Ken Fritts, the engineering services technical manager.
To do this, he needed a clearer picture of the site to easily understand project volumes, work progress and productivity.
The solution:
Previous reality capture tools could not quickly and accurately obtain the relevant data required by the heavy construction industry. REVEAL, AGTEK’s heavy construction software solution, used AI to recognise the objects it captured on Goodfellow’s site and clustered them into groups, such as construction vehicles, material stockpiles, ground and vegetation. AI used point clouds collected from drones or laser scanners to automatically classify and remove unwanted elements from the data.
The outcome:
The Goodfellow team quickly removed any unwanted elements to provide a clear picture of the site and captured the site cleanly to help their field and office staff, understand project volumes and work progress. The result was clean, clear, and accurate data delivered quickly and simply.
“Before REVEAL, finding and sorting points from drone flights was tedious and time-consuming. REVEAL is the next step in take-off support software. It automates the sorting and clean-up of drone data in minutes instead of hours,” Ken said.
Automate and organise data with AGTEK REVEAL
City of Klagenfurt uses geospatial AI platform to optimise solar energy
The challenge:
Klagenfurt’s surveying team had a goal to map properties with different land use categories in the capital of Carinthia, Austria, to better understand their usage and assess whether photovoltaic and solar thermal systems could be installed on building surfaces.
The solution:
A digital twin was created using AI in HxDR, a cloud-native platform for geospatial data at any scale and M.App Enterprise that monitors and evaluates software. The AI technology incorporated 19,000 individual images taken during a four-and-a-half-hour flight over the city at an altitude of 1,200 metres to generate the digital twin.
The outcome:
Using AI, reality data was captured and then loaded into HxDR to create a map of the total area of properties. The different land use categories were documented to better understand how properties within the city are utilised. Roofs generating substantial solar energy were shown in 3D with a colour scale that indicated the level of energy yielded. The digital twin even calculated the effect of the shadows cast by each tree, both in summer and winter, to give a precise calculation of the solar energy potential of each home and business in Klagenfurt. This information is now publicly available in an online portal that was created by the Solar Potential Cadaster of Klagenfurt to help residents evaluate the need for installing a solar power system on their property.
Innovate using geospatial data with HxDR
R-evolution targets 30% more electricity efficiency with solar AI
“The limit now is our own imagination, where before the limit was the technology. We’re going to see these systems doing things and reaching spaces that we couldn’t reach before, anything from emergency response to what we’re doing in green tech.”
- Erik Josefsson, President of R-evolution
The challenge:
In 2021, Hexagon announced the creation of a new subsidiary, R-evolution, to use its range of technology resources and talent to help save the planet.
The first phase focused on solar power, alongside ecological monitoring, and saw R-evolution’s President, Erik Josefsson, acquire a 40-hectare photovoltaic (PV) solar farm site in Archidona, southwestern Spain.
"Our target is quite simple,” he said. “An ambition to deliver electricity at 30% higher efficiency than equivalent farms that do not use Hexagon’s technologies.”
The solution:
AI is used throughout the product range installed at Archidona.
For reality capture and planning, the point cloud data generated from the BLK2GO and RTC360 scanners is automatically registered and meshed via AI-powered workflows.
For monitoring and forecasting, Nexus empowers R-evolution’s solar park through applied AI-powered Internet of Things (IoT) sensors, integration and enterprise application capabilities across assets within the park. It acquires, structures and serves data across databases, API’s and SCADA systems into one mobile application.
OxBlue and BLK247 provide AI video analytics for construction monitoring and site surveillance.
For design, engineering and simulation, Hexagon’s multiphysics Computational Fluid Dynamics solution, Cradle CFD, leverages design and simulation in real time. This software optimises thermal and fluid performance on the solar plant, simulates wind, thermal and energy usage and supports weather forecasting for accurate spot price trading.
The outcome:
‘The simulation was operationalised with AI-based reduced-order modelling and combined with IoT data to produce operational digital twins, that are used to inform and optimise the facility’s operation’, said Erik.
“R-evolution can now see operational performance down to the smallest detail on their mobile devices. By doing so, R-evolution [has established a blueprint to gain operational insight into a solar park's key performance indicators.”
This usage of all available simulation and sensor data to inform better solar farm business decision-making around supply and demand and asset management is crucial for performance improvements.
The photovoltaic site is largely autonomous today and Hexagon’s technologies accelerate this move towards facility autonomy, using data-driven systems to control more functions of the solar farm.