Octave’s Mission: Building Systems That Don’t Break

At Hexagon’s recent global leadership event Hexagon LIVE, Mattias Stenberg, who was appointed in October 2024 to lead Octave, a strategic spinout from Hexagon, delivered a keynote that was both grounded and far-reaching. His talk offered a clear-eyed assessment of the state of critical infrastructure today, and what needs to change if we want our most essential systems to function reliably in the future.

Stenberg, who has been president of Hexagon’s Asset Lifecycle intelligence division since 2017, wasn’t there to announce a product line or unveil another dashboard. His message was more fundamental: we’ve inherited systems built to break. Octave’s mission is to build ones that don’t.

A New Kind of Intelligence for the Real World

Octave is not just another brand. It’s designed as an intelligence layer for the infrastructure that keeps society running. Think power grids, manufacturing systems, highways, water networks, emergency operations centers. The kind of systems where failure isn’t just inconvenient, it’s dangerous, costly, or even catastrophic.

At the core of Octave’s approach is a simple shift in mindset: stop reacting, start anticipating. Octave doesn’t aim to flood teams with more data or more disconnected tools. It aims to turn existing data into foresight, to embed real-time, contextual intelligence into the systems we rely on every day.

The name “Octave” itself is no accident. Inspired by music, it reflects a higher level of coordination and elevation, not just making noise, but orchestrating decisions. In practice, that means integrating AI directly into physical systems so they can detect issues, adapt, and respond long before human operators are even aware a problem is coming.

The Problem: Everything Breaks

Stenberg talked directly – “Things break,” he said, invoking the second law of thermodynamics with a hint of humor.

He pointed out examples we all recognize from the news:

• The Francis Scott Key Bridge collapse, a structural failure that could have been prevented with early detection.
• The Deepwater Horizon disaster, caused by system blind spots and poor integration between safety layers.

He pointed out that in each case, the data existed. The systems failed because they weren’t connected in the right way and lacked the intelligence to interpret early signals.

Octave’s Vision: Failure Is Not Inevitable

Too many organizations accept breakdowns as part of doing business. Delays, downtime, asset failure, they’ve been normalized. Octave’s entire premise is to challenge that mindset.

If systems can sense changes in their environment, process those signals intelligently, and act on them in time, then many failures can be avoided. But that level of performance requires embedded, context-aware intelligence, not generic AI models running in a separate system. Octave’s agents aren’t passive tools waiting for prompts. They’re active participants in system behavior.

Already, Hexagon has started deploying this approach across sectors, from manufacturing and logistics to public safety and energy. The early returns, he stated, are promising: less downtime, better coordination, and more resilient performance.

The Digital Transformation Disconnect

To underline the need for a new approach, Stenberg shared data from a recent C-suite survey conducted by Hexagon’s Asset Lifecycle Intelligence division,:

• Only 1 in 5 companies say they are realizing the full value of their digital transformation investments.
• 76% report using more tools and dashboards than ever, but feel less aligned and less in control.
• One executive summed it up: “More dashboards. More complexity. Less actual visibility.”

This reflects a deeper truth. Digital transformation alone doesn’t guarantee better performance. In fact, if not managed carefully, it can add noise, create silos, and obscure decision-making.

What Octave is offering is a reset: cut the clutter, connect what matters, and turn data into decisions that move the needle.

AI: Less Hype, More Usefulness

Stenberg then discussed the current AI hype cycle.

Yes, AI holds enormous potential, but most AI systems today are untrained, detached from operations, and built in isolation. They’re promising demos in a lab, not solutions in the field.

Octave takes the opposite path. Its AI agents are trained for specific environments, tied into operational systems, and continuously learning. These aren’t showpieces, they’re in the loop, helping machines and humans respond to signals in real time.

The Mirror World: Not Just a Twin, But a Partner

The term “digital twin” has been used in nearly every industry over the past few years. But as Stenberg pointed out, many of these twins are little more than static visualizations, nice to look at, but lacking predictive value.

Octave, Stenberg said, will be redefining the concept. In its vision, the Mirror World is a living, learning model of a physical system, constantly updated, deeply embedded, and able to act. If your digital twin can’t help you see trouble coming, it’s just a digital museum.

The goal is to spot patterns early, detect small signals before they become big problems, and support decisions that prevent, not just mitigate, failure.

Voices from the Field: What Must Not Break

Stenberg’s keynote also featured a panel of customers speaking directly about the pressures they face. Each one brought a unique perspective, but all centered on resilience and visibility.

Donald Lhoest (Carmuese): Global logistics operations require consistent, trustworthy intelligence. When teams are remote and distributed, “data needs to empower, not isolate, them.”

Colonel Mark Shelley (Lee County Sheriff’s Office): “Public safety depends on integrated, real-time intelligence. There’s no margin for error.”

Wade McNabb (Lixil): Factory downtime is a major risk. But human error is just as dangerous. “We need better training systems, and better insight into how our customers actually use our products.”

Joe Bonnet (Worley): “We run 10,000 projects at once. Complexity itself is the problem. If we can eliminate failure points, we can focus on innovation.”

Proof in Practice: The Öresund Bridge

To bring the message home, Stenberg closed with a case study:

The Öresund Bridge, connecting Sweden and Denmark, is equipped with a real-time digital twin. The system detected subtle vibration patterns, signals no human would have caught on their own.

Engineers investigated and discovered microcracks forming. Intervening early prevented what could have been a major infrastructure failure. That’s not theory. That’s the Mirror World in action.

Final Message: Time to Rethink What We Accept

Stenberg’s final words were a challenge to the industry.

“We’ve inherited systems that were built to break. Octave’s job is to build systems that adapt, and don’t fail silently.”

It is not about layering more dashboards or adding complexity. It’s about designing smarter from the start, systems that can think, respond, and evolve.

The post Stop Reacting, Start Anticipating: Hexagon’s Octave Vision for a Smarter and More Resilient Industrial and Commercial Infrastructure appeared first on Logistics Viewpoints.

In the past, tracking a shipping container across continents or monitoring the temperature of a pharmaceutical package in a rural warehouse came with trade-offs: cost, power drain, or unreliable coverage. Asset visibility was reserved for high-value goods, while the rest of the supply chain operated on estimates, paper trails, and phone calls.

This is changing. Two cellular technologies—LTE-M and NB-IoT—are now reshaping long-distance asset tracking. Designed specifically for low-power, wide-area connectivity, they are not flashy or fast, but they are practical. And that practicality is unlocking a new standard of visibility across logistics networks.

A Functional Divide: What Makes LTE-M and NB-IoT Different?

Both LTE-M (Long Term Evolution for Machines) and NB-IoT (Narrowband Internet of Things) were developed under the 3GPP standard. They’re not general-purpose wireless technologies. They’re built for a narrow job: to allow simple devices to send small data packages across long distances with minimal power usage.

Still, they serve different purposes.

LTE-M supports voice (VoLTE), real-time mobility, and bandwidth up to 1.4 MHz. That makes it suited for assets in motion—trucks, railcars, shipping containers. Devices stay connected as they cross cell towers, even at highway speeds.

NB-IoT, on the other hand, is optimized for stationary or slow-moving assets. It operates on just 180 kHz of bandwidth, uses even less power than LTE-M, and excels at indoor or underground penetration. That makes it ideal for warehouse environments, shipping pallets, or cold storage units.

In both cases, devices can last up to 10 years on a battery, waking only to transmit data at defined intervals.

On the Ground: Where These Technologies Are Already Working

In Germany, Deutsche Telekom uses NB-IoT to track reusable transport packaging—low-value but often misplaced. By tagging these items with NB-IoT sensors, they can recover more assets and reduce losses.

In the United States, Roambee uses LTE-M to track pharmaceutical shipments. Their sensors capture not just GPS data, but also temperature, humidity, and light exposure—essential data for compliance and quality control.
Reference

Meanwhile, Sierra Wireless now Semtech offers LTE-M modules built into trackers used across North American freight networks. They enable cross-border asset visibility without requiring complicated roaming workarounds.
Reference

These are operational deployments—quietly streamlining real supply chains today.

From Fragmented Data to Structured Insight

For decades, asset tracking systems lived in silos—fleet telematics in one system, warehouse sensors in another, handheld barcode scans in a third. LTE-M and NB-IoT allow these devices to transmit consistent, time-stamped data that can be ingested directly into ERP, WMS, or TMS platforms.

This leads to several operational changes:

• Shipment ETAs can now be calculated using real-time location data.
• Route deviations, temperature excursions, or tampering events can trigger alerts instantly.
• Idle asset time, loss rates, and turn rates can be tracked quantitatively rather than through assumptions.

These technologies give the supply chain a memory. Not just precise locations—but how long they’ve been there, under what conditions, and whether any anomalies occured.

Strategic Implications: More Than Better Tracking

The shift to LTE-M and NB-IoT is not just about technical improvement. It represents a change in how companies define what is worth tracking.

1. Lowering the Threshold

When connectivity was expensive and battery life short, only high-value goods justified GPS trackers. LPWA (low power wide area) tech brings that threshold down. Companies can now afford to track plastic pallets, returnable containers, or temperature-sensitive packaging. These items were once considered disposable or unmonitored; now, they are part of the data flow.

2. Preparing for 2G/3G Sunset

Many tracking systems still rely on 2G or 3G cellular modules. As networks phase out legacy services, companies are being forced to choose a replacement. LTE-M and NB-IoT offer a forward-compatible path that avoids the higher costs of full LTE or 5G broadband connections.

3. Shifting from Status to Intelligence

In the past, knowing an asset’s last known location was sufficient. Today, organizations are using real-time sensor data to move from reactive to predictive operations. Cold chain breaches, delivery delays, or maintenance needs can be identified before they create service failures or lost revenue.

4. Expanding Global Logistics

LTE-M supports broader international roaming than NB-IoT, but both are becoming more accessible globally as carriers standardize their infrastructure. For multinational operations, this means fewer gaps and less complexity in deploying one global asset tracking framework.

Constraints and Deployment Realities

These technologies are not without limitations:

• Roaming for NB-IoT is still fragmented, limiting its use in cross-border applications.
• Latency for NB-IoT can be high, making it unsuitable for urgent alerts or rapid two-way communication.
• Device provisioning and firmware updates must be managed remotely, especially for assets deployed in hard-to-reach areas.

As a result, organizations must carefully evaluate their device requirements, data latency tolerance, and regional coverage before choosing between LTE-M and NB-IoT.

Supply chains are being redefined not by high-profile innovations, but by infrastructure technologies like LTE-M and NB-IoT that enable quiet, scalable change. These technologies do not replace the need for planning or coordination, but they reduce uncertainty. They give supply chain managers access to verified data instead of estimates. They allow decisions to be made with more context, and fewer assumptions.

In practical terms, LTE-M and NB-IoT make it feasible to track every pallet, every package, and every trailer—not just those deemed high-value. That’s not a breakthrough, it is how resilient, modern supply chains operate today.

The post How LTE-M and NB-IoT Are Revolutionizing Asset Tracking in Global Supply Chains appeared first on Logistics Viewpoints.

Supply chain disruptions have become a persistent operational risk. Geopolitical instability, extreme weather, labor shortages, and fluctuating consumer demand regularly impact global logistics. Traditional supply chain planning, which relies on historical data and reactive adjustments, is no longer adequate for managing these challenges. Artificial intelligence (AI) is reshaping supply chain operations by enabling predictive planning, allowing companies to anticipate disruptions before they occur and adjust operations accordingly.

Amazon is a leader in AI-driven supply chain management. They integrate AI into demand forecasting, inventory optimization, and logistics operations to improve efficiency, reduce costs, and mitigate risks. Let’s examine Amazon’s approach as well as  the limitations of traditional supply chain planning, the operational benefits of AI, and the necessary steps for implementing AI-driven strategies.

Limitations of Traditional Supply Chain Planning

Traditional supply chain planning relies on retrospective analysis. Organizations examine past sales trends, apply seasonal adjustments, and make forecasts based on historical models. When unexpected disruptions occur—a factory shutdown, a shipping delay, or a supply shortage—these models provide little flexibility. Companies must react after the fact, often incurring higher costs and reduced service levels.

A 2023 McKinsey study found that companies relying on reactive supply chain management lose up to 10% of annual revenue due to inefficiencies and missed opportunities. Excess inventory, stockouts, and increased transportation expenses are common consequences of outdated planning methods. Enterprise resource planning (ERP) systems, while effective for tracking transactions and inventory levels, lack the predictive capabilities needed to anticipate and mitigate risks. Executives are left making high-stakes decisions with incomplete information.

AI as a Predictive Tool

AI-driven supply chain planning integrates machine learning, real-time data analytics, and external risk monitoring to anticipate disruptions before they materialize. Unlike static forecasting models, AI continuously refines its predictions as new data flows in. AI systems analyze internal data, such as inventory levels and production schedules, alongside external factors, including weather patterns, geopolitical developments, and consumer sentiment. This enables companies to adjust sourcing, production, and logistics well in advance of potential disruptions.

Amazon’s AI-Driven Supply Chain Planning

Amazon has integrated AI throughout its supply chain to improve demand forecasting, logistics, and inventory management. The company’s AI models analyze sales trends, social media activity, economic indicators, and weather patterns to predict demand fluctuations. This system allows for dynamic inventory adjustments across warehouses, reducing stockouts and minimizing excess inventory.

AI-driven logistics optimization has resulted in faster and more cost-effective deliveries. Dynamic route planning adjusts in real time based on traffic conditions and weather disruptions. Load balancing algorithms ensure efficient distribution across Amazon’s logistics network, preventing bottlenecks and improving delivery reliability.

During the COVID-19 pandemic, Amazon leveraged its AI models to reallocate resources, adjust inventory levels, and reroute shipments in response to shifting demand. The company’s AI-driven supply chain adjustments enabled it to maintain service levels while many competitors faced severe disruptions.

Operational Benefits of AI-Driven Supply Chain Planning

Cost Reduction

AI enables cost reductions by optimizing inventory management, logistics, and procurement. Traditional inventory systems often lead to overstocking, which ties up capital, or understocking, which results in lost sales. AI-based demand forecasting minimizes excess inventory while ensuring sufficient supply. AI-powered logistics optimization reduces transportation inefficiencies by identifying cost-effective shipping routes. Automated warehouse operations streamline order fulfillment, reducing dependency on manual labor. AI-driven procurement tools analyze pricing trends and supplier performance to negotiate better contract terms. Predictive maintenance of transportation fleets reduces downtime and repair costs. AI-enhanced quality control prevents defective goods from reaching distribution networks, minimizing waste. AI fraud detection systems identify anomalies in procurement and payment processes, reducing financial losses.

Demand Forecasting Accuracy

AI models improve demand forecasting by incorporating real-time market data and external variables. Traditional forecasting methods rely primarily on past performance and cannot adapt to sudden shifts in consumer behavior or supply chain conditions. AI integrates external data sources such as weather forecasts, geopolitical events, and social media trends to refine demand projections. AI models continuously adjust their predictions based on evolving market conditions, increasing accuracy over time. This reduces excess inventory while maintaining service levels. AI-powered forecasting allows businesses to identify emerging trends earlier, enabling proactive production planning. Regional demand variations can be anticipated, optimizing inventory allocation across different markets. AI enhances supplier coordination by aligning raw material procurement with production needs. Companies using AI-based demand forecasting lower inventory holding costs while improving order fulfillment rates.

Risk Mitigation

AI enhances risk management by identifying potential supply chain disruptions before they escalate. AI-driven supplier risk assessments monitor financial stability, historical performance, and geopolitical exposure, allowing for early intervention. AI detects logistical risks, such as weather-related transportation delays, and suggests alternative shipping routes. Automated regulatory compliance monitoring ensures adherence to evolving trade laws and import/export restrictions. AI fraud detection tools identify anomalies in transactions, preventing financial losses. Predictive analytics in manufacturing detect potential equipment failures, reducing production downtime. AI-based workforce management tools predict labor shortages and optimize staffing levels. AI cybersecurity applications protect digital supply chain infrastructure from cyber threats. AI-driven risk modeling helps organizations develop contingency plans based on various disruption scenarios. Companies implementing AI-driven risk mitigation strategies recover from disruptions faster and with lower financial impact.

Efficiency Gains

AI improves supply chain efficiency by streamlining processes across procurement, manufacturing, and logistics. Predictive analytics optimize raw material procurement, reducing waste and improving production flow. AI-powered robotics in warehouses increase picking accuracy, reducing mis-shipments and returns. Automated inventory tracking ensures high-demand products are readily available, minimizing stockouts. AI-driven transportation management adjusts delivery routes in real time, optimizing fuel efficiency and reducing transit times. AI-powered quality control detects defects earlier in the production cycle, minimizing waste and rework costs. Digital twins allow companies to simulate different supply chain scenarios before making operational adjustments. AI-driven chatbots handle supplier negotiations, freeing procurement teams to focus on strategic planning. AI-powered invoice processing reduces errors and processing delays in financial transactions. AI-based supply chain simulations improve strategic decision-making by testing different operational models before implementation.

Regulatory and ESG Compliance

AI enhances regulatory compliance and sustainability tracking by automating data collection and reporting. AI-driven emissions monitoring systems track carbon output from transportation and manufacturing, ensuring compliance with environmental regulations. AI verifies ethical sourcing practices by analyzing supplier labor conditions and identifying potential human rights violations. AI and blockchain integration improve supply chain transparency, enabling better traceability of goods from production to distribution. AI automates compliance reporting, reducing administrative burden and improving audit readiness. AI-based logistics optimization minimizes fuel consumption, aligning with corporate sustainability objectives. AI-enhanced waste management identifies opportunities for material recycling and reuse. AI-powered predictive modeling helps organizations prepare for upcoming regulatory changes, reducing non-compliance risks. Organizations integrating AI into sustainability initiatives improve investor confidence by demonstrating proactive ESG compliance.

Implementation Considerations

Executives considering AI adoption must first assess their data infrastructure. AI-driven models require standardized, high-quality data across all supply chain functions. Organizations should prioritize high-impact use cases, such as demand forecasting and supplier risk assessment, before scaling AI implementation. AI adoption requires investment in talent with expertise in machine learning, data analytics, and supply chain management. Selecting the right AI solutions is critical—tools must be scalable, compatible with existing systems, and industry-specific. Measuring AI performance through defined KPIs ensures continuous improvement and accountability.

Challenges and Constraints

AI adoption presents several challenges. Data quality remains a common issue—without accurate inputs, AI predictions are unreliable. Organizational resistance to AI-driven decision-making can slow implementation, requiring executive leadership to drive adoption. Initial AI deployment costs can be high, but efficiency gains and cost reductions typically offset expenses within 12 to 18 months. Over-reliance on AI models without human oversight can lead to unintended operational risks.

Amazon’s AI-driven supply chain demonstrates the operational benefits of predictive planning. AI enhances demand forecasting, logistics optimization, risk mitigation, and regulatory compliance. Organizations that fail to adopt AI-driven supply chain planning will face continued inefficiencies and competitive disadvantages. The transition from reactive to predictive supply chain management is no longer an option—it is an operational necessity.

The post Amazon and the Shift to AI-Driven Supply Chain Planning appeared first on Logistics Viewpoints.