Publications
:
The integration of artificial intelligence in ERP systems represents one of the most significant technological shifts in business software today.
If you work in manufacturing, you’re probably familiar with the challenge of avoiding defective products. A defect as small as a missing or unidentifiable barcode can result in downtime and disappointed customers. To limit the number of defective products, manufacturing companies struggle to manage these visual quality inspections. This article will introduce how to improve these inspections through computer vision for quality control – a field of artificial intelligence that trains computers to interpret and understand the visual world. Problem – the cost of quality inspections Every company in manufacturing has the challenge of reducing the number of defective products. Quality inspection in manufacturing is a must; otherwise, faulty products will appear one way or another. Defective products can cause damage and result in unexpected expenses, leading to customer complaints, downtime, labor costs, and scrapped products. Therefore, dedicated people are often used as the quality checkpoint, visually looking at the production line – which costs time and money. The quality check is critical, but manual visual quality inspection slows down the production phase since you must ensure the inspector can keep up with the production line. Is it worth it? Well, it’s necessary to ensure high-quality products – even though it’s a costly expense for the company. In some cases, where risk is high, the cost of letting defective products slip through is considered worse. E.g., delivering faulty products to a client/customer could result in lost contracts/agreements. Therefore, it’s essential to avoid these scenarios and reduce errors, even if that often means having a lot of resources in place and a slower production phase. A well-designed systematic quality inspection will have a positive effect on: Downtime Defective product Loss of revenue Lost customers Wasted time Wasted resources & man-hours Wasted money Decreased OEE / utilization And more.. In most cases, posting a person on the production line at each point of quality inspection is often too expensive – so let’s talk about alternatives using new technologies and solutions like computer vision. Solution – reducing defects and human errors by using computer vision analysis The solution is to install relatively inexpensive cameras in locations where you typically place, or would like to place, a person for visual inspections. By using digital images and video from these cameras, we can train computer vision models to perform analysis. These models enable the cameras to accurately identify and classify objects during production line inspections. An edge computing device processes each frame, performs the analysis, and ultimately outputs the result that the model was trained to generate. With these frames from a video, the model can be trained within a few hours to identify defects in real-time wherever you have repetitive quality inspections. We can then use this data to react to what the camera “sees.” Example 1: Best by dates on packaging Implementing object detection and quality control with computer vision on a packaging production line to detect valid printed Best By dates on packaging. This will allow the detection of misprinted, invalid, or missing Best By codes to be removed from production before being boxed and sent […]
1. INTRO The utilization of AI technology is one of Novacura’s strategic themes. We are a high-technology company delivering advanced software products, so we can’t imagine ourselves without AI. We aim to build an affordable bridge to AI for our customers and transform an unfamiliar expert tool into a daily companion. We had the opportunity to speak with our CEO, Johan Melander, who explained Novacura’s approach and strategy in this area. This is a part of the interview that was performed during the Novacura Flow User Conference on 2024-09-18, between: Johan Melander (JM) CEO of Novacura, and Łukasz Majer (LM), Business Solutions & Marketing Director at Novacura (the host). 2. Where AI Can Help Us: Key Areas of Application LM: Johan, we’ve had many opportunities to discuss the role of AI in today’s technological landscape. You mentioned that AI has become somewhat of a buzzword these days. I would like to explore this area further so it doesn’t remain an “empty” slogan. When I think about AI utilization in the context of Flow, I can imagine its support in the following areas: Flow developer co-pilot, helping developers build apps End-user co-pilot – guiding users through the process and automating certain steps for them Process optimization co-pilot – a tool that consumes telemetry data (process stats), helps detect anomalies (fraud), and identifies bottlenecks in processes Now, where do you see the biggest potential? Which of these directions will be Novacura’s priority – what’s on our roadmap? JM: Well, all these areas are actually on our radar, and we address them in our roadmap. Of course, not everything will be available from day one, but you’ve touched on all important aspects of AI utilization in our platform and solutions. We are currently developing the developer co-pilot as part of Flow Studio. We also have some AI-based automations embedded in our apps (like invoice recognition, receipt analysis, and computer vision object detection). 2.1. AI USE CASES: FOR FLOW DEVELOPERS LM: Could you then quickly explain what AI could provide in relation to these three categories of AI companions? Let’s start with the developer companion, perhaps? JM: Sure, but I want to clarify that I’m now presenting our strategic vision, not necessarily the elements you’ll see in one of the next releases. I can imagine AI helping Flow developers in the following areas: Flow Script creation – You say something in natural language, and AI creates a Flow script program; Data import companion – You show your Excel/CSV/XML file, and the AI tool prepares the import script, maps objects to your internal data structures, etc.; Data layer creation – You type something like, “I need objects to store my invoices with details” and AI creates objects for the invoice header and invoice lines, establishing all relations with other existing objects like Customer, Supplier, or Project. Importantly, it already knows which properties are required to represent this object type properly; User interface creator – You describe your needs for the UI, and the tool builds it for you. You may also point to a data object […]
Introduction Novacura participated for the second consecutive year in one of Poland’s largest logistics and manufacturing events – Tomorrow Logistics Conference in Poznan (March 13th). This year, our focus was on presenting an informative case study regarding the implementation of the ‘Just in Sequence’ Strategy in one of a major furniture manufacturing company, a key subcontractor for a leading furniture producer. The event provided an excellent opportunity to engage with over 100 industry experts and professionals while showcasing our low-code solution, Novacura Flow. Attendees also benefited from insightful presentations both on stage and talks at various booths throughout the event. One notable presentation was delivered by Łukasz Majer, Business Solutions & Marketing Director at Novacura. He shared a compelling customers case titled ‘Implementing the Just in Sequence Strategy in a Large Company – A Successful Case Study,’ highlighting practical insights and successful strategies. PRESENTATION SUMMARY Łukasz presented a case study of a major furniture manufacturing company, a key subcontractor for a leading furniture producer, showcasing the successful implementation of the “Just in Sequence” strategy, an advancement of Just in Time. With the initial problem being limited pre-production warehouse space, he explained the five key success factors that enabled the customer to implement this complex and difficult-to-achieve way of working: All steps in the Just in Sequence supply process must be digitized and supported by a mobile app. Employees responsible for managing the applications should have a solid business background. Optimization must encompass the entire process, rather than focusing solely on one specific area or division. The person responsible for implementing the change must support the process from the initial analysis stage to post-go-live support. Implemented solutions must undergo continuous improvement, with the underlying technology (such as low-code) supporting this iterative process. Our speaker demonstrated how Novacura Flow, a low-code solution, facilitates the seamless operation of all applications on any given device automatically. Łukasz also emphasized that this solution aligns with industry trends, as highlighted during the expert debate: “The Bright and Dark Sides of Digital Transformation – How to Digitize an Organization Wisely and Integrate Systems,” stressing the necessity of focusing on open technologies that allow for modification and integration. The presentation is available in Polish: If you are interested in the broader interview with the client, go to the link: (in Polish): NOVACURA PODCAST ABOUT THE EVENT The Tomorrow Logistics Conference held in March, organized by the Logistics Manager Magazine team, served as a platform for innovative discussions on logistics and supply chain management. Novacura actively participated in the substantive discussions, led by Łukasz Majer and Tomasz Czerwiński (Business Solutions Architect). Our team also showcased our solutions at the booth, engaging in insightful conversations with potential clients who demonstrated a solid understanding of their needs in manufacturing, logistics, and production. We also offered a small gift to our booth visitors, using Lego sets as an analogy to highlight the flexibility and freedom in building solutions with our tool. Main Presenters: Piotr Susz (Locura): Discussed blockchain for supply chain traceability in Poland. Marek Kuropieska (Aspekt) and Ewa Wardak-Turkiewicz (Zebra Technologies): Highlighted flexible and swift automation processes. […]
Most companies, even the most digital ones, have a paper trail embedded in their processes. These pieces of paper can be used in the field by technicians, in the back office, or even at the administrative desks. A large amount of time goes into preparing templates, printing documents, filling in forms, all just to have someone type them back into the system of record – if one exists. We’ve talked to the Director of Product Management Labs at Novacura, Paul Phillips, about his experiences and problems and potential solutions such as Form Recognizer, related to this type of work – relevant for all companies who utilize paper-based processes. Problem Paper trails are cumbersome, costly, and prone to errors no matter which person is executing the work. I’ve seen Job Safety Analysis forms in the field where the same form duplicated for the same job dozens of times. Instead of growing on the knowledge of the field technicians, each time the same job was performed, a brand new JSA was filled out causing loss in efficiency. Another example of this is during the pandemic, COVID vaccinations are required, I’ve seen back-office administration taking days-weeks documenting vaccination records of employees. This manual job can result in: Loss in efficiency Increased cost Poor data in the system of record (“fat finger” of data) All these processes could be digitally transformed to significantly streamline the process. But digitizing a document to enter in a computer screen isn’t the only way to do it. Why does this problem exist? The quickest way to get any new process in place that requires documentation is to have a user fill out a form. The quickest way to get the form out to all the employees is to create a template in Word or Excel and have everyone print out paper copies. It is typically more time-consuming to digitize the process, but as we know this will streamline the data capture downstream. We often try to standardize a process through a document but don’t always take into consideration the amount of data we lose as an implication of not considering the data entry part of every process. How can we measure the problem? The best way to measure the problem is to describe an example and Paul describes: “In one of my old companies, we had just gone live with an ERP system, and we were trying to convert Standard operating procedures (SOPs) to Standard Jobs in our ERP system. To do this, we hired a temporary data admin to help us get the paper processes into a digital system. We brought this person on for 3 months and ended up hiring them full-time to key information in from paper to the ERP because of how time-consuming it was. Arguably we could’ve hired 2 more people to get it done quicker but it wasn’t in the budget. What came out of the process wasn’t the highest quality – due to the admins’ lack of process knowledge, came at a higher-than-expected cost and also had errors – as expected for any person who enters data all day long. At that […]