# Sentispec — full content export > AI for Forwarding, Distribution and Warehousing. Full text of all case studies and blog posts, prepared for LLM ingestion. Last generated: 2026-09-01T09:40:50.774Z --- # Case Studies ## S&P 500 Global Manufacturer — Inspector · Manufacturing URL: https://sentispec.com/case-studies/sp500-manufacturer/ A Fortune 500 manufacturer uses Inspector for real-time container fill rate visibility, taking the guesswork out of load planning. ### Background A global 3PL services this S&P 500 heavy equipment manufacturer, managing a steady flow of 85+ weekly containers moving product across their international logistics network. ### Challenge - Average load fill of just 43%, with over half of each container shipping as air. - No transparency over trailer utilisation across weekly volumes. - Costs of 85+ weekly containers with no visibility or tools to drive performance improvements. ### Outcome - Consolidated product across loads, delivering €1M annual savings. - Identified under-utilised loads where additional freight could be added without extra carrier cost. - Replaced forecasts with actual data, unlocking more reliable optimisation scenarios. - Reduced CO2 footprint per shipped unit. ### Metrics - 43%: Starting average load fill - €1M: Annual savings delivered - 85+: Weekly containers monitored --- ## Top 10 Global Automotive OEM — Inspector · Automotive URL: https://sentispec.com/case-studies/top10-automotive/ One of the world's largest automotive manufacturers monitors loading quality on inbound and outbound freight, cutting transit damage claims. ### Background Sentispec services this global logistics provider's transportation network for a top 10 global automotive manufacturer: a Europe-wide network running up to 5,000 daily departures. ### Challenge - Average load fill across the network sat at just 46%. - With 5,000 daily departures, no one could answer the basic question: is 5,000 the right number? - No real-time feedback loop to connect loading quality to cost, utilisation, or sustainability KPIs. ### Outcome - Real-time load data replaces estimation: every departure is measured. - Network-level opportunities to consolidate and reduce daily departures. - Reduced operational cost per shipped vehicle and component. - Reduced CO2 footprint across the European network. ### Metrics - 5,000: Daily departures monitored - 46%: Starting load fill --- ## Global Wind Turbine Manufacturer — Inventory · Manufacturing URL: https://sentispec.com/case-studies/global-manufacturer-inventory/ A global wind turbine manufacturer replaced manual stocktaking with Sentispec Inventory, protecting uptime worth thousands per lost minute. ### Background This global wind turbine manufacturer operates several production sites worldwide. Each facility holds local stock to keep production lines running continuously and to protect the profitability of expensive equipment. ### Challenge - Low inventory accuracy was causing production downtime. - Every minute of production downtime costs €7,000, so the business needs inventory accuracy of 99%+ to protect output. - Manual stocktaking could not keep pace with the accuracy demands of high-value, high-tempo manufacturing. ### Outcome - Sentispec Inventory delivers significantly higher accuracy in stock levels and stock locations. - Less time wasted retrieving stock to production. - Reduced equipment downtime and fewer production bottlenecks, directly protecting equipment profitability. ### Metrics - 99%+: Inventory accuracy target met - €7,000: Cost per minute of downtime avoided --- ## Global Shipping & Logistics Company — Handler · Logistics URL: https://sentispec.com/case-studies/global-logistics-handler/ A global 3PL deployed Handler at key distribution hubs, saving 7 minutes per load by removing manual admin and scanning on every container. ### Background This global shipping and logistics company operates a network of warehouses servicing some of the world's largest retailers, with high quality demands and very high throughput. ### Challenge - High quality demands from demanding enterprise customers. - Very high container throughput across the warehouse network. - Manual admin and scanning work on every container, burning operator time on every load. ### Outcome - 6% warehouse productivity improvement across the network. - 7 minutes of loading admin saved per container. - Increased effective shipping capacity without adding headcount. - Higher customer satisfaction from demanding retail clients. ### Metrics - 7 min: Admin time saved per container - 6%: Warehouse productivity gain --- ## Global Retailer — Inventory · Retail URL: https://sentispec.com/case-studies/global-retailer-inventory/ A global 3PL replaced manual stocktaking with Sentispec Inventory: 100,000 pallets scanned monthly, stock accuracy pushed to 99.97%. ### Background This global shipping and logistics company offers warehousing and distribution services to a range of retailers and manufacturers across Northern Europe. Many of their high-profile clients require frequent, comprehensive stock audits of 30,000 to 100,000 pallets. ### Challenge - Stocktaking was a manual, labour-intensive, and costly process. - Retailers and manufacturers demanded accurate stock levels and efficient loading to avoid out-of-stock situations that directly hit revenue. - Frequent audits of 30,000–100,000 pallets were consuming significant labour and time. ### Outcome - Notably higher customer satisfaction across high-profile retail and manufacturing clients. - 5–8% productivity improvement across Northern European warehouses. - 95% reduction in missing-pallet-related labour. - Stock accuracy increased to 99.97%. - 5–10% labour cost reduction across Northern European warehousing. ### Metrics - 100,000: Pallets scanned per month - 99.97%: Stock accuracy achieved - 95%: Reduction in missing-pallet labour ### Customer quote "The solution brings real change and efficiency from day one." — Head of Warehousing and Distribution --- ## Top European Furniture Retailer — Inspector · Retail URL: https://sentispec.com/case-studies/european-furniture-retailer-inspector/ Published: 2026-04-23 A leading European furniture retailer swapped historical-average trailer bookings for real m³/LDM data: 5% transport cost and CO₂ savings. ### Background One of Europe's leading furniture retailers operates a high-frequency outbound logistics network across multiple distribution centres, moving substantial daily volume to stores and direct-to-customer deliveries. ### Challenge - Trailer space booked on historical averages, not actual load data. - No visibility into effective m³ or load-metre utilisation per departure. - Systematic over-ordering of trailer capacity, driving avoidable freight costs and excess CO₂ emissions. ### Outcome - 2% transport cost savings realised within the first month of go-live, growing to 5%. - Trailer space booked on statistically superior load parameters, eliminating guesswork from capacity planning. - Full visibility into m³ and LDM utilisation per departure, replacing estimation with accurate real-time load data. - Reduced CO₂ emissions, directly supporting the retailer's sustainability strategy. ### Metrics - 5%: Transport cost savings - < 1 month: Time to first measurable savings - m³ / LDM: Utilisation visibility per departure --- # Blog Posts ## Logistics at a crossroads: amplifying expertise with intelligent automation URL: https://sentispec.com/blog/logistics-crossroads-amplify-expertise/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-04-17 Source: https://www.linkedin.com/feed/update/urn:li:activity:7450826775556907008/ The logistics industry is at a crossroads. For decades, we've relied on experience, gut instinct, and manual processes to manage transportation. But rising freight costs and tighter margins are forcing a fundamental shift. The companies thriving today aren't just working harder — they're working smarter. Instead of guessing at fill rates, they have real-time visibility. Rather than discovering booking inaccuracies at the dock, they catch them before trucks roll. They've moved from reactive damage claims to proactive load quality monitoring. This isn't about replacing human expertise — it's about amplifying it with intelligent automation that turns your existing cameras into powerful data sources. The question isn't whether this transformation will happen. It's whether you'll lead it or be left behind. What's your biggest challenge with transportation visibility right now? --- ## Freight rates climb, trucks run empty — and the math doesn't add up URL: https://sentispec.com/blog/freight-rates-up-trucks-half-empty/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-04-16 Source: https://www.linkedin.com/feed/update/urn:li:activity:7450502139971399680/ Freight rates climbing 12% while trucks run 15% empty. The math doesn't add up, but the pattern is clear: poor capacity planning is costing the industry billions. Most logistics teams are flying blind on actual fill rates. They book based on estimates, hope for the best, and pay premium rates when loads don't fit as planned. The hidden cost isn't just the wasted space — it's the cascade effect. Missed consolidation opportunities. Last-minute spot market bookings. Carriers passing inefficiency costs back to shippers. Real-time fill rate data changes this equation entirely. When you can see actual utilization as trucks load, you can make smarter decisions about consolidation, routing, and carrier selection. The technology exists today. The question is: how long can your operation afford to plan in the dark? --- ## The $184B cost of supply chain visibility gaps URL: https://sentispec.com/blog/184b-visibility-gap/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-04-15 Source: https://www.linkedin.com/feed/update/urn:li:activity:7449381142169808897/ $184 billion. That's what supply chain disruptions cost companies in 2024 alone. The biggest culprit? Lack of real-time visibility into what's actually happening at loading docks. While we debate AI strategies in boardrooms, trucks are leaving half-empty, containers are being damaged without documentation, and transport teams are flying blind on actual fill rates. The math is brutal: if you're paying for capacity you're not using, you're bleeding money every single day. The good news? The cameras are already there. The loading bays exist. The data is waiting to be captured. What's missing is turning that visual information into actionable intelligence that prevents the $184B problem from repeating in 2026. How much visibility do you really have into your transport utilisation right now? --- ## The gap between theory and practice in supply chain optimisation URL: https://sentispec.com/blog/gap-theory-practice-supply-chain/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-02-23 Source: https://www.linkedin.com/posts/aclaudi_supplychainoptimization-logisticstech-inventorymanagement-activity-7431615578785828864-wWhf Just read a fascinating academic paper from Chalmers University that dives deep into supply chain optimization challenges. While the research is highly technical, it reinforces something we see daily in our work with global logistics companies. The paper highlights how complex mathematical models can solve optimization problems, but here's what it doesn't address: the real-world gap between theoretical solutions and practical implementation. In our experience working with Fortune 500 manufacturers and logistics giants, the biggest challenge isn't finding the optimal solution — it's getting accurate, real-time data to feed into these models. You can have the most sophisticated optimization algorithm in the world, but if your inventory data is off by even 10%, your results become meaningless. This is where technologies like computer vision are game-changing. Instead of relying on manual counts or outdated systems, companies can now capture precise dimensional and inventory data automatically. The result? Those beautiful optimization models actually work in practice, not just on paper. The research community is pushing the boundaries of what's mathematically possible. Now it's up to us in industry to bridge that gap with practical, implementable solutions. What's been your experience with implementing optimization models? Where do you see the biggest gaps between theory and practice? --- ## The logistics industry is at a fascinating crossroads URL: https://sentispec.com/blog/logistics-industry-fascinating-crossroads/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-02-20 Source: https://www.linkedin.com/posts/aclaudi_logisticstechnology-supplychainoptimization-activity-7430521739035537408-QTC_ The logistics industry is at a fascinating crossroads. While AI promises revolutionary change, the reality on the warehouse floor often tells a different story. Here's what we're seeing work in practice: 🔍 Real-time visibility beats prediction algorithms — Most supply chain disruptions stem from not knowing what's happening right now, not from failing to predict the future. 📦 Dimensional accuracy drives bottom-line results — One client reduced shipping costs by 18% simply by measuring packages correctly with computer vision. No complex algorithms needed. ⚡ Implementation speed matters more than sophistication — The best AI solution is the one your team actually uses. We've seen simple, well-executed systems outperform complex ones that sit unused. 🤝 Human expertise amplifies AI impact — The most successful deployments combine AI's consistency with human judgment and domain knowledge. The companies winning today aren't necessarily using the most advanced AI. They're using the right AI for their specific challenges and implementing it thoughtfully. What's your experience been? Are you seeing practical AI applications making a real difference in your operations, or are you still navigating the hype versus reality gap? --- ## True value unlocks through process understanding URL: https://sentispec.com/blog/value-unlocks-through-process-understanding/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-02-20 Source: https://www.linkedin.com/posts/aclaudi_transportcost-ai-supplychainvisibility-activity-7430529999612522497-7D_G The true unlocking of value comes from a deep understanding of the working process. In this case our customer and we both ended up with an intimate understanding of the inner workings of their transport booking — truly an enjoyable experience. --- ## Episode V – Return of the Data URL: https://sentispec.com/blog/episode-5-return-of-the-data/ Author: Ivan Markov, Head of Commercial, Sentispec Published: 2026-02-20 Source: https://www.linkedin.com/posts/ivankmarkov_transportcost-ai-supplychainvisibility-activity-7430527459856384000-wxaB Episode V – Return of the Data This is where the story becomes operational. Using automated data capture, we replaced a static assumption with a dynamic Conversion Ratio: - Updated daily - Calculated per store - Applied across 200+ stores with weekly deliveries For the first time, planners had a decision they could trust — grounded in real, current data. And the impact was immediate: ✅ Goods left on the warehouse floor disappeared ✅ Overbooking of load meters stopped ✅ Transport costs reflected reality, not theory The result? €1.3 million in annual savings. No heroics. No guesswork. No blind faith. Just visibility, translated into action. Sometimes, the Force isn't AI. It's finally seeing what's been there all along. --- ## Sometimes the biggest wins come from solving the smallest inefficiencies URL: https://sentispec.com/blog/biggest-wins-smallest-inefficiencies/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-02-19 Source: https://www.linkedin.com/posts/aclaudi_supplychain-logisticsoptimization-warehouseproductivity-activity-7430170546400661504-r4BD A global logistics company faced manual bottlenecks consuming 7 minutes per container for admin and scanning work. After implementing Sentispec Handler, they achieved: - 6% warehouse productivity improvement - 7 minutes saved per container in loading admin - Increased shipping capacity without additional resources - Higher customer satisfaction scores Small inefficiencies compound into meaningful business impact when addressed. What operational bottlenecks are draining time in your operations? --- ## Container optimisation: beyond 'fill and ship' URL: https://sentispec.com/blog/container-optimisation-beyond-fill-and-ship/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-02-19 Source: https://www.linkedin.com/posts/aclaudi_supplychain-logistics-containeroptimization-activity-7430237568610283520-u4p9 Kuehne+Nagel's latest insights on container optimization highlight a critical reality: the traditional approach of "fill and ship" is no longer enough in today's supply chain landscape. Their emphasis on data-driven container utilization resonates deeply with what we're seeing across industries. But here's what caught my attention — while optimizing space is crucial, the real game-changer lies in real-time visibility throughout the entire process. The challenge isn't just knowing how much space you have; it's understanding: • What's actually in each container vs. what the manifest says • How efficiently that space is being used across different product categories • Where optimization opportunities exist before goods even reach the dock We're working with logistics leaders who've discovered that accurate, automated measurement and verification can improve container utilization by 15-25% while reducing costly surprises at destination. The future of container optimization isn't just about better planning — it's about having the right data at the right time to make smarter decisions throughout the entire supply chain. What's been your biggest challenge with container optimization? Are you seeing similar opportunities in your operations? --- ## Episode IV – The Missing Piece URL: https://sentispec.com/blog/episode-4-missing-piece/ Author: Ivan Markov, Head of Commercial, Sentispec Published: 2026-02-19 Source: https://www.linkedin.com/posts/ivankmarkov_transportcost-ai-supplychainvisibility-activity-7430189721621983233-Wjn7 Episode IV – The Missing Piece We now knew where money was leaking. We knew how much could be saved. But we still couldn't answer the most important question: 👉 How do you actually realise these potential savings? So we did what most data projects never do. We stopped analysing. And started listening. Together, we ran a series of workshops, walking through the retailer's entire logistics planning process — from store ordering all the way to transport booking. That's when we found it. The Conversion Ratio (CR). Stores ordered in cubic meters. Those cubic meters were converted into load meters using a ratio. That ratio determined transport bookings. The problem? - It was static - Based on legacy data - Never questioned - Never updated The galaxy had changed. The ratio had not. This was the missing link between insight and action. --- ## Episode II – The Proof of Concept Strikes Back URL: https://sentispec.com/blog/episode-2-proof-of-concept/ Author: Ivan Markov, Head of Commercial, Sentispec Published: 2026-02-17 Source: https://www.linkedin.com/posts/ivankmarkov_transportcost-ai-supplychainvisibility-activity-7429501664396713984-Wm2G Episode II – The Proof of Concept Strikes Back We began where all rebellions begin: small, fast, and focused. Using our Inspector solution, the retailer's team started capturing images of outbound trucks with a simple app. No heavy infrastructure. No long implementation. From day one, the data spoke. What had been booked in load meters versus what was actually used in reality differed significantly. And not occasionally. Consistently. Very quickly, a pattern emerged: - Load meters were systematically overbooked - Transport costs were being calculated on assumptions, not reality - Capacity was being paid for but not used - Goods were being left on the floor, not reaching the stores After just 200 data points, the numbers told a compelling story: 👉 A potential saving of €700,000 to €1.4 million per year But the rebels knew better than to celebrate too early. The insight was powerful. The process was not. --- ## Route planning: the art and science of modern logistics URL: https://sentispec.com/blog/route-planning-art-and-science/ Author: Andreas Claudi, CEO, Sentispec Published: 2026-02-16 Source: https://www.linkedin.com/posts/aclaudi_supplychain-logistics-routeoptimization-activity-7429097525589999616-zlv4 Route planning has evolved far beyond simple A-to-B navigation, and this recent article from NetworkON captures why it's become such a critical differentiator in modern logistics. What resonates most is their emphasis on the 'art and science' approach. While algorithms can optimize distances and fuel consumption, the real magic happens when you combine data intelligence with practical understanding of operational nuances. Here's what I see missing from most route planning discussions: 🔍 Real-time visibility gaps — Routes are only as good as your ability to adapt when reality diverges from the plan. Unexpected delays, capacity changes, or quality issues can derail even the most optimized route. 📊 Data integration challenges — Route optimization works best when it can access comprehensive, real-time data about inventory levels, vehicle capacity utilization, and actual delivery conditions. 💡 The human factor — Driver expertise, customer relationships, and local knowledge often trump pure algorithmic optimization. The article rightly points out that companies excelling in route planning gain competitive advantage through better service at lower costs. But in my experience, the real winners are those who can seamlessly blend route optimization with end-to-end supply chain visibility. What's your biggest route planning challenge? Is it the initial optimization, real-time adaptation, or something else entirely? --- ## Episode I – A New Approach URL: https://sentispec.com/blog/episode-1-new-approach/ Author: Ivan Markov, Head of Commercial, Sentispec Published: 2026-02-16 Source: https://www.linkedin.com/posts/ivankmarkov_transportcost-ai-supplychainvisibility-activity-7429102606578253824-qzah Episode I – A New Approach A long time ago, in a Nordic supply chain not so far away… Every logistics transformation starts with disputing a popular belief. Most of the time, that belief is: 'This is just how things are done.' When we first approached a Nordic furniture retailer, we didn't come with a pitch deck full of features. We came with curiosity. Transport costs were increasing. High volumes of trailers were being booked weeks in advance to service 200+ stores. Yet no one could confidently say whether those bookings reflected reality on the dock. The question wasn't who was wrong. The question was simpler — and more powerful: 👉 Are we overpaying for transport without even realising it? Together, we agreed on a shared mission: - No assumptions - No historic spreadsheets - No opinions Only ground truth from the warehouse floor. This wasn't about technology yet. It was about visibility. And so, Episode I ended — not with answers, but with alignment. --- ## Cutting CO₂ through AI-driven trailer utilisation URL: https://sentispec.com/blog/cutting-co2-ai-trailer-utilisation/ Author: Linda Moens, VP Commercial, DHL Global Forwarding Published: 2025-07-29 Source: https://www.linkedin.com/posts/linda-moens-b381786_sustainability-gogreen-logistics-activity-7355856821636456448-8vW7/ 🌍 At DHL, cutting CO₂ isn't just a goal — it's a strategy. What's one powerful way to reduce emissions right now? 📊 Use AI to make every trailer count. Thanks to our collaboration with Sentispec, we're leveraging real-time space utilisation monitoring to maximise trailer loads, eliminate empty miles, and slash avoidable emissions — all without compromising service. ♻️ Smarter use of space. 🚛 Fewer wasted trips. 🌱 A cleaner supply chain. This is how we're driving sustainability and value — one shipment at a time. --- ## Smarter loading equals smarter logistics URL: https://sentispec.com/blog/smarter-loading-smarter-logistics/ Author: Linda Moens, VP Commercial, DHL Global Forwarding Published: 2025-07-22 Source: https://www.linkedin.com/posts/linda-moens-b381786_logistics-supplychainoptimisation-dhlglobal-activity-7353346219156381697-7NLX/ At DHL, we believe smarter loading = smarter logistics. What if AI could tell you exactly how well your containers are packed — and how to do it better? With Sentispec's load analysis, we're boosting container fill quality, cutting down on empty space and damage risk, and saving real money for our customers — all while delivering the same high level of service. 📦 Less waste. 📉 Lower costs. 🚚 Smarter transport. 🔍 Curious how we're redefining value in forwarding services? --- ## Sentispec and DHL: Unlocking AI-powered container intelligence URL: https://sentispec.com/blog/sentispec-dhl-container-intelligence/ Author: Andreas Claudi, CEO, Sentispec Published: 2025-07-15 Source: https://www.linkedin.com/feed/update/urn:li:activity:7350790949456756737/ So stoked to finally lift the veil on some amazing stuff we are doing with DHL! It's a massive pleasure working with such a forward thinking, customer-centric company, to unlock truly game changing capabilities in transport asset utilisation using AI. Ultimately this allows DHL to deliver best-in-class value for money for their customers' supply chain needs. --- ## How to scale predictable AI URL: https://sentispec.com/blog/how-to-scale-predictable-ai/ Author: Andreas Claudi, CEO, Sentispec Published: 2024-01-23 Source: https://www.linkedin.com/pulse/how-scale-predictable-ai-andreas-claudi-mw7if/ The use of AI in logistics can be cumbersome to scale because of the need for extensive training and huge data loads. But by prioritising only needed information and ignoring the rest, we can reduce a 10 MB sized problem into a 0.5-1 MB instead, meaning vastly reduced cost of training and operation as well as increased scalability. As most realise by now, AI is way wider than just Generative (which in itself is built on Machine Learning AI, which is built on Neural Networks AI), and encompasses everything from Computer Vision, Image Manipulation, and Robotics, to such exotic topics as Swarm Intelligence. From an observer's point of view, the common trait amongst all of these AI technologies is that they appear at times to act non-deterministically — that is to say that from time to time they will react in an unexpected manner on certain stimuli — be it a written text or an image observed. The underlying technologies are, by the way, largely deterministic, however the appearance of seemingly random behaviours is caused by the variance of inputs, which speaks to the real challenge of most Machine Learning type AI: "AI can only react with 100% predictability on previously seen input data." What happens when environments change? Now, when we are trying to deploy a scalable solution across a wide range of potentially dynamic environments, the above statement is sure to spell disaster. The most prolific examples of this are to be found in the fields of Self Driving Cars and Autonomous Robotics. Self Driving Cars Consider a car AI which has been trained to operate in California, where the weather is relatively consistent, and you take that same AI and ask it to drive a car in the middle of a winter snowstorm in Finland. Result? It's pretty obvious — 5 minutes and it ends up in a ditch. The solution? Train the car to operate in every climate — simple? How many types of traffic cones are there in the world? We need to recognise them. How about traffic signals? What about different types of clothing for people? Dog? Cat? Goats? And here is the problem — achieving 100% predictability just on the input analysis is a task of infinite size, and we haven't even started discussing how to achieve adequate reasoning and behaviour based on those input stimuli. Autonomous Robotics Robotics have been, and are being, deployed for automation tasks at scale in manufacturing, pharmaceuticals, retail, and logistics at scale — but they suffer from a similar problem when moving to full autonomy — in particular around the input stimuli when mobile around a physical facility. The problem is that every facility is different in some respects, which means that to achieve full autonomy, specific training in that facility is necessary. Furthermore, every time the physical environment is changed, for example by moving a shelf system or production line, the system needs to be retrained, and whilst recalibrating, the system cannot function with a very high level of predictability. That means that as a particular autonomous system scales to a multitude of facilities, this again becomes a problem of infinite complexity, thus requiring an infinite amount of computing power to achieve 100% predictability. Is there a solution? Yes, and no. Achieving 100% predictability on anything based on Machine Learning, with any kind of complicated data is prohibitively expensive. The good news is that in most cases it is not necessary. Typically, any complex AI is based on a hierarchy of reasoning on top of the input analysis layer. When those layers are made robust towards fluctuations in input analysis results, we can — just like humans do by the way — compensate, and still make reasonable conclusions and thus determine good courses of action. Simultaneously, there is a trick to working with complex data, which is to reduce the information contained in the input data set — simply put we filter out irrelevant data to the point that we extract very simple conclusions from highly complex data. Our Sentispec Inspector solution extracts volumetric fill rates of trailers based on a single image of the trailer contents. In doing so, we filter out everything outside of the trailer as the first step, because it is irrelevant in our determination of the fill rate inside. This in itself reduces a 10 MB sized problem into a 5 MB problem. On top of that, we don't need 4K resolution so we downscale it by a factor of 10 — which means that instead of dealing with 5 MB of information we are now looking only at 0.5-1 MB instead. As we can do these operations with minimal loss of accuracy, we achieve two things: 1) Vastly reduced cost of training and operation. 2) Increased scalability as the scope of our training is reduced to only the key factors in the images. "The key to success with scalable AI is in determining what information is really needed, and ignore the rest — not unlike what the human brain does." --- ## The disappointment of autonomous AI URL: https://sentispec.com/blog/disappointment-of-autonomous-ai/ Author: Andreas Claudi, CEO, Sentispec Published: 2022-12-13 Source: https://www.linkedin.com/feed/update/urn:li:activity:7008328860542869504/ I've been mulling this one over for a while, and wanted to see how much help I could get from ChatGPT to write a decent article (about 30% of the content is ChatGPT — not great, not terrible, saved a bunch of work). --- ## End-to-end supply chain visibility URL: https://sentispec.com/blog/end-to-end-supply-chain-visibility/ Author: Andreas Claudi, CEO, Sentispec Published: 2022-09-06 Source: https://www.linkedin.com/pulse/end-end-supply-chain-visibility-andreas-claudi/ End-to-end supply chain visibility represents the holy grail of logistics for any manufacturer or retailer, though achieving it remains extremely challenging due to the numerous partners involved in global transportation networks and costly identification touchpoints. The key benefits of supply chain visibility include measurable KPIs, competitive advantage, forecasting capabilities, customer satisfaction, and compliance. However, proposed solutions like RFID and barcode scanning portals require substantial hardware investments and industry-wide adoption, making them impractical at global scale. Many touchpoints already use barcode scanning, but the associated costs often burden shippers. Computer vision offers a cost-effective alternative, leveraging existing cameras in mobile devices, security systems, and standard forklift-mounted cameras to track items by barcode from source to destination. Sentispec Access converts visual data into real-time tracking and business data, enriching supply chain systems (WMS, TMS, ERP) while improving operational economics. Data vision is quite simply visionary for business. --- ## Musings on human vs. computer vision URL: https://sentispec.com/blog/musings-human-vs-computer-vision/ Author: Andreas Claudi, CEO, Sentispec Published: 2022-07-26 Source: https://www.linkedin.com/pulse/musings-human-vs-computer-vision-andreas-claudi/ When I founded Sentispec, it was from the basic notion that in the animal kingdom, the one sense that really sets humans apart is our eyes. Not because our eyes are mechanically superior, but because of the immense processing power positioned immediately behind the eyes — our human brain. Human brains are incredibly powerful — even our fastest super computers are still 1000 times less powerful (and are not at all mobile). As such this begs the question — what are the manual processes which one might improve if just one could observe them 24/7? Whereas this would be unviable using human labour, it might just be possible using computer vision. Simply place a camera to observe the manual process and give you back the statistical results. And thus Sentispec was born. In the field of AI, the area which has always held the most potential data is computer vision, and it is only now, with good and cost effective cameras and processing power, that we can truly start tapping into the power of it, by placing cameras in the right places and realtime processing the results into business insights. Take for example our Automated Stock Taking solution — we generate some 200–300 MB of data per second, which would have been unthinkable to process in real time at an affordable cost just a few years ago. Stock taking is a great example of a tedious manual process which can be optimised with computer vision in a quick affordable way, and with a Return on Investment within 3 months. If you run a pallet hotel of, say 100,000 pallets, you will likely count these 4 times per year at a manual productivity of 50–100 pallets/hour. With a labour cost of €25/hour this cost is around €100–200,000 annually. With the Sentispec AST solution we max out north of 3000 pallets/hour. Yes. 3000. The short version of this is an 80–95% cost saving, or even better, the ability to free up several full time people from cost-only work and reallocate them towards revenue generation. Simultaneously, you get a higher level of stock transparency and spend less time looking for lost pallets. And "all" we did was put a high speed camera on a pallet and a forklift, and process the data with AI. Beyond Stock Taking, we have solutions available for ensuring Loading quality, and optimising Fill Rates across an entire network. We are looking at solutions for automatically scanning goods out at gate (saving 2–3 minutes per trailer load), and generating performance and productivity metrics for Picking operations. That's probably enough rambling for one sitting so let me end on the immortal words of Captain Jean-Luc Picard: "To boldly go where no one has gone before!" ---