Insights
Field notes from a decade of industrial AI — what it takes to climb the levels, written by the people building the road.
THE SERIES
The Road to Level 5
The scale, documented — one chapter per level, written as we build the road.
- 0CHAPTER 0In progress
- 1CHAPTER 1The Industrial AI Playbook
- 2CHAPTER 2AI Assistants for Operators: Setting the Foundation for AI-Driven Industrial Operations
- 3CHAPTER 3One Year Closer to the Autonomous Factory
- 4CHAPTER 4In progress
- 5CHAPTER 5In progress
New to the series? Start with the framing piece — Towards the Sentient and Autonomous Factory
TYPE
LEVEL
106 PIECES
- THE ROAD TO LEVEL 5 · CHAPTER 3One Year Closer to the Autonomous FactoryWhat changed in AI, what didn't, and what industrial autonomy actually requires: a one-year retrospective — three eras of AI in under a year, why the bottleneck is now operational rather than technical, and how plants move from Level 2 to supervised autonomy. LEVEL 3
- THE ROAD TO LEVEL 5 · CHAPTER 1The Industrial AI PlaybookA cookbook for turning AI into operational advantage in the asset-heavy industries: ten claims for making analytics and AI an operational capability, each illustrated by a chlor-vinyl energy-optimization deployment. LEVEL 1
- Towards the Sentient and Autonomous FactoryWhat a cup of coffee can teach us about AI-driven manufacturing: four co-evolving AI systems (Assist, Self-Monitor, Strategize, Act) as the blueprint for the factory that knows itself and runs itself. LEVEL 5
- THE ROAD TO LEVEL 5 · CHAPTER 2AI Assistants for Operators: Setting the Foundation for AI-Driven Industrial OperationsUsing large language models and AI agent systems to give operations teams on-demand, collaborative, trustworthy insights from industrial data. LEVEL 2
- Building agentic AI on Arundo FoundationWhy a separate graph-based domain model plus time-series storage and a compute layer is the right architecture for agentic AI in heavy-asset industries, with a natural-gas shipping walkthrough. LEVEL 2
- Navigating Sustainability: The Path to Scope 3 Emissions Management for Maritime LeadersRegulatory and market drivers of Scope 3 disclosure in shipping, and CarbonPath as the maritime-specific management solution.
- Leveraging Technology for ESG Compliance: How Maritime Companies Can Measure, Manage, and Report Scope 3 EmissionsWhy Scope 3 emissions are hard for maritime companies and how Arundo CarbonPath addresses measurement, management, and reporting.
- Arundo CarbonPath - Scope 3 emissions management for shippingTurn Scope 3 compliance into business value: how CarbonPath connects the shipping value chain to measure and manage supplier emissions.
- The Fundamentals of Adopting Asset Reliability SoftwareA buyer's framing for asset-reliability software: business outcomes, build vs buy, and integrating insights into existing workflows. LEVEL 1
- Maximizing Asset Performance with AI-Powered Tools: Direct Outcomes at ScaleMarathon's configurable workflows highlighted by IoT Analytics as a leading example of moving from AI-driven alerts to prescriptive outcomes. LEVEL 1
- Arundo Analytics Launches Energy Optimization Application for Chlorine ElectrolysisPress release: launch of the chlorine-electrolysis energy-optimization application for INEOS Inovyn, with 3-5% typical electricity cost savings. LEVEL 1
- Revolutionizing heavy asset maintenance: The power of LLM-based virtual assistants (whitepaper)Whitepaper edition of the LLM virtual-maintenance-assistant study: question answering and information retrieval over tanker maintenance logs, RAG, and fine-tuning. LEVEL 2
- Minimize energy costs with the Energy OptimizerOptimizing the chlor-alkali process against spot electricity prices: 3-5% realizable electricity cost savings, with an INEOS Inovyn case study. LEVEL 1
- Value Creation from Advanced Analytics in the LNG IndustryOpportunities for LNG asset owners in applying advanced analytics to equipment data: BOG handling, reliquefaction efficiency, GCU failure prevention. LEVEL 1
- Revolutionizing Heavy Asset Maintenance: The Power of LLM-based Virtual AssistantsA hands-on guide to building LLM-based virtual maintenance assistants: question answering and information retrieval over real maintenance logs, plus RAG and fine-tuning. LEVEL 2
- Arundo awared \"Most promising Data Analytics Solution Provider\"Arundo is named Most Promising Data Analytics Solution Provider 2022 by CIO Review.
- Marathon software to smarten Predictive Maintenance in the Oil & Gas IndustryPress release: Dietsmann Smart Labs selects Arundo as its partner for predictive maintenance of industrial plant, using Arundo's Marathon software. LEVEL 1
- The Story of MacGregor OnWatch ScoutHow MacGregor built a condition-monitoring service offering on Arundo technology to lower its customers' total cost of ownership. LEVEL 1
- Cyber Security and Streaming Data: IoT, AI and Security PoliciesHow industrial companies can start with advanced analytics under strict cybersecurity policies without giving up security. LEVEL 0
- Bridging the Digital Divide: Legacy Documents in a Digital WorldWhy legacy paper and image documents slow industrial work, and how to bridge analog and digital information. LEVEL 0
- Digital Transformation in the EPC IndustryHow shifts in the EPC industry, from cost pressure to buying criteria, raise the importance of an overall digital strategy. LEVEL 0
- Monitor Your Critical and Less-Critical Assets With Machine LearningHow machine learning and advanced analytics can complement traditional reliability programs, for critical and less-critical assets alike. LEVEL 1
- Applying ML Across Common Bottlenecks in Industrial Business ProcessesHow to pick machine learning use cases in industrial operations by targeting common bottlenecks across workflows, judged on feasibility and value. LEVEL 1
- Production Optimization in the age of Artificial IntelligenceHow oil and gas production engineers can use digital technologies and data science to manage many more wells with the same or fewer resources. LEVEL 1
- How can I get the Most Value out of my Digital Twin?What a digital twin is, and why starting from narrow use cases and business value gets the most from one. LEVEL 2
- How Many Sensors Do I Need to Get Started With Machine Learning?How many sensors are needed to start monitoring critical assets with machine learning, and where edge analytics fits. LEVEL 0
- Tsaug: An Open-Source Python Package for Time Series AugmentationArundo releases tsaug, an open-source Python package that augments time series data to train machine learning models. LEVEL 1
- Getting SaaS-y About Industrial Data ProductsGuiding principles for delivering SaaS-style industrial data products despite complex, non-standard asset configurations. LEVEL 1
- What Your SCADA System Isn't Telling YouHow operators can turn the large volume of SCADA data into meaningful decisions. LEVEL 1
- How Arundo Switched to Docs as CodeHow Arundo's documentation team moved to a Docs as Code publishing process using existing tools.
- Machine Learning for Industrial Use Cases: Hard-Won LessonsLessons from delivering data and machine learning solutions to customers across industrial use cases. LEVEL 1
- ADTK: Open-Source Time Series Anomaly Detection in PythonArundo releases ADTK, an open-source Python toolkit for time series anomaly detection, built from its data science project experience. LEVEL 1
- An Overview: Technical Terms Used in Digital TransformationDefinitions of technical terms used in digital transformation in heavy-asset industries, from advanced analytics to anomaly detection. LEVEL 0
- Another great year at the Rice Data Science ConferenceRecap of the 2019 Rice Data Science Conference: themes and trends, and the poster Arundo presented. LEVEL 1
- Getting Started with Machine Learning for CompressorsHow to get started with machine learning on compressors to reduce downtime. LEVEL 1
- Data Science in Oil & Gas: Observations & Thoughts from ATCE 2019Recap of ATCE 2019 in Calgary: talks on equipment monitoring, predictive maintenance and time-series deep learning. LEVEL 1
- UXDX and Design Systems - Observations and Thoughts From UXDX 2019Notes from Arundo's design team on design systems and user experience, from the UXDX conference in Dublin.
- Equipment Monitoring and Alarm Fatigue in the Age of AIWhy more alerts do not help operators already facing alarm fatigue, and how to pick the right problem for equipment monitoring. LEVEL 1
- VIDEO: From POC to Production in Predictive MaintenanceVideo transcript: challenges heavy-asset industries face deploying AI in predictive maintenance, and tips for moving from proof of concept to production. LEVEL 1
- Enterprise Kubernetes Summit Houston 2019 - ReviewReview of the Enterprise Kubernetes Summit in Houston: Rancher, GitLab, Datadog and Aqua talks.
- What's Missing From Asset-Heavy Digitalization?Why digitization efforts that focus on data, sensors and equipment often miss the user. LEVEL 0
- Smart Iron: Industrial OEMs and the Internet Of ThingsCapabilities industrial OEMs need to build IoT-enabled products and services. LEVEL 0
- Why is Security so Important in Industrial Internet of Things?Why security matters in industrial IoT, and the main points Arundo regularly discusses with customers. LEVEL 0
- Machine Learning in Industrial Applications: Research to RealityProblems that arise when moving machine learning from research and development into industrial use, and a human-in-the-loop approach to handle them. LEVEL 1
- The Executive Guide To the Industrial Internet of ThingsAn executive introduction to the IIoT, its use cases, and implementation challenges. LEVEL 0
- Pump Analytics: Solving Pump Inefficiency and CO2 EmissionsWhy pump inefficiency contributes to CO2 emissions, and how advanced analytics gives insight into pump performance. LEVEL 1
- On the Edge: The Increasing Importance of Decentralized Analytics in IIoTWhy edge computing and decentralized analytics help IIoT companies avoid unplanned equipment downtime. LEVEL 0
- How to Start Thinking About Architecture for Data-Driven SolutionsArundo's view on the architecture questions customers raise early when starting data-driven solutions, and how a solution fits into existing infrastructure. LEVEL 0
- Edge Analytics and the Industrial Internet Of ThingsHow to run analytics on lighter-weight local computers when equipment is in remote or moving locations. LEVEL 0
- Predictive Equipment Maintenance - Anomaly Detection for Industrial OperationsHow to start an equipment-analytics program with cluster-based anomaly detection when historical failure data is scarce. LEVEL 1
- Why do so many AI Programmes Fail?Why many corporate digital and AI programmes fail to deliver results, despite large investments in data science Centres of Excellence. LEVEL 1
- 4 Things You Should Know: Starting an Industrial Data Science InitiativeFour things to know before starting an industrial data science initiative: what problems it can address, and how to approach it. LEVEL 1
- More Pilots Than an Airline: Scaling AI and Digital TransformationCommon pitfalls as companies move from proof of concept to value at scale with IoT and machine learning, and what to do about them. LEVEL 1
- Why is Industrial IoT Implementation so Hard?Why industrial IoT projects struggle to deliver value and what to do about it. LEVEL 0
- What's Edge Analytics?How edge analytics models extend edge technology beyond data acquisition in an IIoT solution. LEVEL 1
- Implementing IIoT - 6 Critical Factors you Need to ConsiderSix factors to consider when implementing an Industrial Internet of Things solution, starting with value. LEVEL 0
- What You Should (and Shouldn't) Expect From Your CDOWhat industrial leaders should and should not expect from Chief Digital Officers, and where their focus should be. LEVEL 0
- Creating Business Value from Unstructured DataWhy most operational data from an industrial asset goes unused, and how to extract value from unstructured data. LEVEL 1
- Digital Operations, Industrial Services, & the Next Wave of Operational ImprovementArundo COO Stuart Morstead on past waves of operational performance improvement and the role of data-driven insights in the next one. LEVEL 1
- From Local Controls to Cloud Analytics: Tags to Bridge OT & ITHow tag-based data contracts cut the time and effort needed to turn local operating data into business value. LEVEL 0
- How Anomaly Detection is Changing Industrial OperationsWhat anomaly detection is, how it may apply to industrial operations, and the concepts and constraints to know. LEVEL 1
- Rapid IoT Projects in Industry - Where Should You Focus?Where to focus when setting a strategy for data analytics and IoT projects in an industrial environment. LEVEL 0
- What You Need to Know About Digital Wells and Field OptimizationHow digital technologies help oil and gas production managers keep more wells producing at their potential with the same or fewer resources. LEVEL 1
- Data Science Applications: 5 Things an Operations Engineer Should KnowFive things operations engineers should know about data science applications, from the human in the loop to false positives and false negatives. LEVEL 1
- The Journey to Predictive Equipment MaintenanceThe path to predictive equipment maintenance, including common challenges and pitfalls. LEVEL 1
- From Industrial Data to Business Value - Why Data isn’t the New OilWhy most industrial analytics initiatives stall after the pilot, and the one capability to invest in to succeed. LEVEL 1
- DevOpsDays Houston 2019 - ReviewTakeaways from DevOpsDays Houston 2019: running Kubernetes, product teams and DevOps as a process.
- Design Thinking Meets Heavy Assets - How Design can Unpack Problems in a few HoursHow design thinking helps asset-heavy industries frame problems and reach solution building in a few planned hours. LEVEL 0
- Google Next: Chaos Engineering (aka Game Days)Notes from a Google Next 2019 talk on chaos engineering and how engineering teams can run game days.
- De-mystifying the Terminology of Digital TransformationDefinitions of IIoT, big data, data science and AI, to separate lasting technology shifts from jargon. LEVEL 0
- What is Edge Computing - Everything you Need to KnowWhat edge computing is and how it gives access to data in rugged, remote or disconnected locations. LEVEL 0
- Arundo Analytics and Dell Technologies Launch IoT Bundle for Oil & Gas and MaritimePress release: Arundo and Dell Technologies launch an IoT bundle pairing Dell Edge Gateway devices with Arundo Edge Agent software for maritime and oil and gas. LEVEL 0
- Arundo Analytics Expands Global Presence, Adds Prominent Industry Leaders to Board of DirectorsPress release: Arundo adds Stuart Morstead, Tore Myrholt and Thomas Malone to its Board of Directors.
- Arundo Analytics Announces Successful Deployment at Global Industrial Thread Manufacturer CoatsPress release: Arundo deploys software at thread maker Coats to collect machine signals and report on machine utilization and equipment effectiveness. LEVEL 1
- Arundo Analytics Named a Cool Vendor in Gartner 2018 Cool Vendors in IoT Analytics ReportPress release: Gartner names Arundo a 2018 Cool Vendor in IoT Analytics.
- Arundo Analytics Unites with DNV GL’s Open Industry Platform Veracity to Stream Data Analytics for MaritimePress release: Arundo makes its analytics software available on DNV GL's Veracity platform for vessel and floating-installation operators. LEVEL 1
- Arundo Analytics Expands into StockholmPress release: Arundo expands into Stockholm under Martin Lundqvist to serve the Nordic market, and plans to host a Maintenance Analytics Summit.
- Arundo Analytics Partners with Acteon Group to Deliver Advanced Machine Learning Applications in Subsea ServicesPress release: Arundo and Acteon Group partner to deliver machine learning models for subsea applications. LEVEL 1
- ABB delivers virtual flow meters powered by Arundo AnalyticsPress release: ABB and Arundo build the first cloud-based virtual multiphase flow meters for offshore oil and gas, as part of the ABB Ability portfolio. LEVEL 1
- WorleyParsons and Arundo Analytics to deliver machine learning to the energy and resources industryPress release: WorleyParsons, through Advisian Digital, and Arundo Analytics announce a collaboration to deliver machine learning to the energy and resources industry. LEVEL 1
- Arundo Analytics Launches Industrial Equipment Condition & Performance Monitoring ApplicationPress release: Arundo announces its Condition & Performance Monitoring application (CPM) for industrial equipment, demonstrated with Fuglesangs. LEVEL 1
- Arundo Analytics Raises $25M Series A Funding to Bring Large-Scale Machine Learning Into Asset-Heavy IndustriesPress release: Arundo announces an initial closing of $25 million on its Series A financing round.
- Arundo Analytics Announces Latest Release of Industrial IoT Software SuitePress release: Arundo announces its Fall 2017 general availability release, with upgrades to its Edge Agent, Composer and Fabric software. LEVEL 0
- Arundo Analytics Brings Silicon Valley Insights To Heavy IndustryPress release: Arundo announces cloud-based, edge-enabled software for advanced analytics in heavy industry. LEVEL 1
- Arundo Analytics Named to MIT STEX25Press release: Arundo is named to the MIT STEX25 group of startup companies.
- SICPA Partners with Arundo Analytics to Enable End-to-End Compliance Monitoring of Government Natural Resources RegulationsSICPA Partners with Arundo AnalyticsPress release: SICPA and Arundo announce an exclusive partnership on data-driven compliance monitoring of hydrocarbon resources for governments. LEVEL 1
- Smart Iron: Industrial OEMs and the Internet of ThingsPress release: Arundo COO Stuart Morstead gives a keynote at the Wells Fargo Energy Conference on machine learning and data science in the energy industry. LEVEL 1
- Data Science in HoustonA data scientist on returning to Houston and watching its data science community grow.
- Get Insight and see Photos of How Arundo Works and Plays as a Team at the Arundo OffsiteArundo staff from Oslo, Houston and Palo Alto meet in Texas for an offsite week and internal hackathon.
- Arundites in ParisAccount of a one-week workshop at the GE Digital Foundry in Paris, building a power scheduling application on the Predix platform. LEVEL 1
- Ellie Dobson, Director EMEA Data Science, gives a talk at local Meetup during visit to Houston in JanuaryEllie Dobson recounts speaking at the Houston Data Science meetup and her path from particle physics to data science at Arundo.
- Microsoft, Carnival Maritime, and Arundo team up to test the waters with machine learning in the Cortana Intelligence SuitePress release: Arundo uses the Microsoft Cortana Intelligence Suite and machine learning to analyze water challenges on Carnival Maritime ships. LEVEL 1
- Fuglesangs And Arundo Agree To Deliver The Digitized Pumping Experience Through Cloud-Enabled Machine LearningPress release: Fuglesangs signs a Letter of Intent with Arundo to enable its pumps with the Arundo Q solution. LEVEL 1
- Arundo participates in the 2017 Geilo Winter School in eScience organized by SINTEFFirst-person account of hosting a session on machine learning in industrial settings at the 2017 Geilo Winter School in eScience. LEVEL 1
- Happy Holidays from ArundoHoliday note: Arundo moves into new offices at Wergelandsveien 7 in Oslo.
- Norwegian Entrepreneurs Demand Focus on Entrepreneurship from Norwegian GovernmentNews: Arundo CEO Tor Jakob Ramsøy joins Norwegian entrepreneurs urging the government to invest in startups.
- Arundo to attend the 10th McKinsey Innovation Forum in Singapore on October 25, 2016Press release: Arundo CEO Tor Jakob Ramsoy presents at the 10th McKinsey Innovation Forum in Singapore on October 25, 2016.
- Arundo Analytics Inc and Statoil ASA Enter into Agreement to Establish Data Science / Big Data Based Optimization of OperationsPress release: Statoil awards Arundo a contract as preferred vendor for data science and analytics capabilities. LEVEL 1
- Plug and Play selects Arundo as 1 of 24 companies to join its 5th batch of IoT startupsNews: Plug and Play selects Arundo as one of 24 companies in its fifth batch of IoT startups.
- Ellie Dobson, Director of Data Science, quoted in Nature, the International Weekly Journal of Science.Media mention: Arundo's director of data science is quoted in Nature on the need to understand how AI systems reach conclusions.
- Houston’s Arundo Taps Into the Industrial Internet for Energy FirmsMedia mention: Xconomy interviews Arundo COO Stuart Morstead on using historical data and predictive models to avoid unplanned downtime. LEVEL 1
- UPDATED - Tor Jakob Ramsøy, CEO of Arundo was recently interviewed by Agnes Grutle Brande, senior manager at Sprint ConsultingIn Norwegian, with English subtitles: Sprint Consulting interviews Arundo CEO Tor Jakob Ramsøy on the startup story and direction.
- Arundo Raises Additional Capital and Secures Contract with Key ClientNews: Arundo secures NOK40 million in financing and a major contract with Statoil, as reported by Dagens Næringsliv. The linked source is in Norwegian.
- The Industrial Internet of Things has really taken off!News: Arundo's CEO Tor Jakob Ramsøy is interviewed by Kapital, a Norwegian business magazine, about how Arundo started and how it presents business use cases.
- Arundo: Simple Design for Complex SituationsArundo launches a refreshed website and begins a brand transition, with the design thinking behind it.
- Brillio Expands Capability in Big Data Analytics for Asset Intensive Industries with Arundo InvestmentPress release: Brillio announces an investment in and partnership with Arundo to expand big data analytics for asset-intensive industries.