IFOY Audit
Finalists undergo the three-stage IFOY Audit as part of the IFOY Audit, which takes place during the TEST CAMP INTRALOGISTICS. This consists of the scientific IFOY Innovation Check, the IFOY Test and the jury test.
DeepCargo
Koiotech
DeepCargo is the first AI-based freight measurement system to automatically measure multiple freight pieces and scan barcodes on a moving forklift – without process interruption. This proprietary deep-learning-software measures all materials, scans barcodes, and assigns them automatically. Our innovation eliminates process stops and gives logisticians immediate revenue boost, drastic cost reductions, and accurate master data for data-driven process optimization.
Product name and company
DeepCargo – Koiotech
Description
DeepCargo is a stationary drive-through gate located at central hubs or warehouse receiving gates. It solves the core problem in logistics: a lack of high-quality master data. Up to 50% of all freight dimensions are incorrectly specified or missing, resulting in billions in losses due to incorrect pricing, inefficient truck utilization, and scheduling.
The DeepCargo gate combines Koiotech´s proprietary AI software, state-of-the-art 3D sensor technology, and barcode scanners. When the forklift passes through the gate at up to full process speed (up to 17 km/h), the AI identifies and distinguishes between the forklift, the pallets, and the freight items, precisely measures the freight dimensions (length, width, height) and assigns the correct barcode to each package – even for multi-cargo transports with complex-shaped cargo items. The captured master data (dimensions, barcode, symbols) is transferred in real time to the customer’s warehouse and transport management systems (WMS/TMS). The process flow remains uninterrupted, while the quality of the master data is greatly improved.
Innovations
DeepCargo has achieved a technological breakthrough in freight measurement for intralogistics through the use of artificial intelligence. The market currently offers no satisfactory solutions, as existing systems are fundamentally limited: they are based on analytical algorithms that do not meet the high process requirements of logistics.
These systems cause significant process disruptions: Forklifts have to stop or slow down considerably (often to 2.5 – 7 km/h). They are limited in terms of shapes (often only rectangular), materials (black foil), and freight sizes (no oversized items). In addition, all of them fail when it comes to simultaneously measuring multiple pieces of cargo, including correct barcode assignment.
The result is massive process disruptions: freights need to be singulated, resulting in more trips at reduced speeds and placing additional strain on time-critical handling processes.
DeepCargo is the first and only measurement system based on a proprietary, self-learning AI-technology. This solution meets all industry-relevant requirements for the first time: DeepCargo measures multiple, complex-shaped cargo items of all sizes (including oversized items) and materials (including black foil) simultaneously on the forklift at full speed (up to 17 km/h) with an accuracy of below 3 centimeters.
This dynamic multi-cargo technology not only captures dimensions and barcodes in motion, but is also the first system to assign each barcode precisely to the correct physical cargo item. DeepCargo thus enables the highest master data quality without compromising processes for the first time.
Market relevance
DeepCargo has enormous market relevance, as it addresses a central, previously unsolved core problem in intralogistics. The logistics industry has been searching for a practical solution for over ten years, as existing systems from other providers fail to meet the technical requirements of the process. The market is largely unsaturated, even though global markets in the areas of smart logistics and smart warehousing are growing exponentially, with CAGRs (Compound Annual Growth Rates) of 21.7 % and 14.2 %, respectively.
The demand for such technology is particularly high among the primary target group, less than truckload logistics. The reasons for this are particularly heterogeneous freight volumes, the need for consolidation, and significant uncertainties in master data, which have a direct impact on revenue.With this target group, Koiotech is potentially addressing around 4,000 logistics warehouses in the EU, each of which can use several DeepCargo systems per location.
Secondary target markets are air freight, contract logistics, and companies with their own warehouse logistics (1PL, first-party logistics). The focus of use is on processes that are handled with forklifts and where the freight items have heterogeneous and largely unknown shapes.
Overall, all target groups represent around 50 % of European logistics market.
Customer benefits
The customer benefits of DeepCargo are immediate, measurable, and strategic. This solution transforms warehouse processes through three key levers:
- Immediate revenue increase (revenue capture): DeepCargo closes the critical data gap, as up to 50 % of shipment dimensions are missing or incorrect. Through complete, accurate recording, it can be ensured that every shipment is correctly priced and billed. Pilot projects and industry analyses confirm an average profit growth of 3 to 5 %.
- Drastic cost reduction: The system eliminates manual, error-prone measurement, data entry, and additional time-consuming manual barcode scanning. This reduces the personnel required for these activities by approximately 95%, which corresponds to 2 to 3 full-time equivalents per shift. Expensive misloads are avoided.
- Maximum efficiency & future-proofing: DeepCargo is the only system that maintains the process flow without interruption. Dimensions (L/W/H) and barcodes (including 2D/QR codes) are captured as the cargo passes by and automatically assigned to the correct package by AI. The data, including photo documentation for quality assurance, is immediately available in the WMS/TMS.
DeepCargo provides accurate, real-time digital data. This forms the basis for future AI optimizations (e.g., truck utilization, route planning) and leads to massive cost savings and a significant reduction in environmental impact.
Seamless integration and a fast return on investment (ROI) of typically less than 12 months make DeepCargo a crucial investment for digital transformation.
Test report
DeepCargo is, according to Koiotech, the first stationary cargo measurement system using artificial intelligence.
The system consists of two 3D cameras, Koiotech’s proprietary deep learning software, and multiple barcode scanners. The number of barcode scanners depends on the size and resolution of the barcodes to be read.
As the forklift passes through the gate without slowing down, the AI identifies and distinguishes the truck and the load on the forks. According to the manufacturer, this is possible at a speed of 17 km/h.
The dimensions of the load (length, width, height) are determined with an accuracy of less than three centimeters. At lower truck speeds, the system is even more accurate.
A key innovative aspect of DeepCargo is the use of deep learning in combination with 3D sensor technology and barcode scanners. While other measurement systems rely on analytical algorithms and often face limitations in terms of speed, shape recognition, and material handling, DeepCargo uses self-learning AI to process complex situations. The system can even measure multiple loads simultaneously, complex shapes, and pallets wrapped in black stretch film.
DeepCargo also scans all barcodes, including mixed loads or irregularly shaped cargo. This enables object recognition and data linking in a single integrated process, resulting in a consistent and reliable dataset. Manual post-processing is no longer required.
No mandatory driving path is required for accurate measurement. The gate can be approached from all directions. Diagonal passage is also possible.
The captured data is transmitted in real time to the customer’s Warehouse Management System (WMS) or Transport Management System (TMS). Initial projects are already in live operation, including at Rhenus Logistics in Göttingen.
The improved data flow primarily provides operational efficiency for the customer. With real-time availability and direct integration into existing systems, processes can be better controlled. In addition, activities such as measurement, data entry, and barcode scanning are largely automated. This reduces the risk of human error, and employees can be deployed for other tasks.
IFOY Test verdict
DeepCargo is a technologically advanced solution providing reliable master data. Measuring and identifying cargo is not new, but the combination of AI, real-time data processing, and process integration makes this solution unique. The multi-cargo technology not only captures dimensions and barcodes in a single fast process. DeepCargo also precisely links each barcode to the correct physical cargo. And it does all of this at full speed, without interrupting any processes.
IFOY Innovation Check
Functionality / Type of implementation
The DeepCargo measurement platform from Koiotech reliably detected, measured, and assigned the corresponding barcode to a wide variety of cargo items, including multiple and complex-shaped loads, on a fast-moving forklift during testing. In pilot customer use, successful cargo measurement was also demonstrated on multiple forklifts passing through simultaneously. This functionality is achieved through the combination of multiple RGB-D sensors for monitoring the passage area and the fusion of several state-of-the-art approaches for semantic segmentation in RGB and 3D sensor data. Through image-to-image tracking of segmentation results at ten frames per second and temporal accumulation of the sensor data stream, the final measurement is performed with very high precision within an accuracy range of three centimeters. The system is largely independent of external lighting conditions.
The measurement system integrates seamlessly into existing Warehouse Management Systems (WMS) and complies with relevant regulations such as the AI Act, Cybersecurity Act, and GDPR (through pixelation of individuals). The solution is designed to incorporate operational sensor data for continuous model improvement. In particular, failed evaluations are manually re-annotated, included in model training, and model updates secured by unit tests are deployed to all customers. During the IFOY Audit, a new forklift also had to be quickly trained in this way, which is expected to become less frequent as the data base grows.
Novelty / Innovation
In the market of measurement platforms, DeepCargo is the first and only solution that uses state-of-the-art foundation models for semantic segmentation of cargo items on forklifts and therefore enables simultaneous measurement of multiple, even complex-shaped items at full forklift speed (17 km/h), including correct barcode assignment. Even challenging materials (black films, reflections) and oversized cargo up to truck size, which exceed the camera’s field of view, can be fused across image sequences through the tracking system and processed completely.
Customer benefit
The customer benefit is very high, on the one hand due to a significant increase in data quality from often less than 50 percent to 85 to 90 percent correctly measured cargo items, and on the other hand due to resulting cost savings of three to five percent through correct billing. In addition, up to 95 percent of personnel costs are saved through automated capture compared to manual measurement and entry into the WMS. Due to the consistently cost-efficient hardware design of the system, Return on Investment (ROI) is achieved for customers in less than 12 months, with a maximum of two scan gates per warehouse required. With commissioning within one day and fine-tuning to the new working environment within a few days, the system is quickly operational at customer sites.
Market relevance
DeepCargo addresses an unsolved problem in parcel logistics: measurement and correct assignment to the respective freight data record when multiple packages are on a forklift. Market relevance is very high, as the product was developed in direct response to demand and in collaboration with two pilot users (Rhenus, Zufall Logistics), and successful use has been demonstrated in four paid Proof of Concepts (PoCs). Strong scaling is now being pursued via logistics networks. The addressable market for use in warehouses and transshipment facilities (approximately 4,000 logistics halls in the EU) as well as in secondary applications such as air freight, contract logistics, and companies with in-house warehousing is significant.
IFOY verdict
Koiotech´s highly sophisticated DeepCargo measurement system, implemented with state-of-the-art evaluation methods, enables practical application of automated freight measurement in significantly broader use cases thanks to its market-oriented functionality and learning capability. The significant scaling potential can be rapidly realized through existing pilot customers and logistics networks.
| Functionality / type of implementation | ++ |
| Novelty / Innovation | + |
| Customer benefits | ++ |
| Market relevance | ++ |
| ++ very good / + good / Ø balanced / – less / — not available | |
Videos
Playlist
1 Videos
Playlist
3 Videos
Image/video credits: Koiotech
Discover more
Statutes
In den Statuten sind die Grundwerte und die Standards des IFOY AWARD festgelegt, nach denen die Organisation handelt. Dazu gehören unter anderem Gremien, Wahlomodus, Bewertungskriterien sowie der Code of Conduct.
Committees
The International Intralogistics and Forklift Truck of the Year has three bodies: Jury, sponsors, and advisory board.