Industrial AI: How Smart Technologies are revolutionising confectionery
Confectionery manufacturers are in need of flexible equipment that can generate a greater range of products, with the ability to bring new products to market quickly and with smaller batch runs. This is where the latest in Industry 4.0 concepts and technologies can help suppliers to expand end product portfolios while maintaining quality and safety for the consumer, reports Daisy Phillipson
Consumer expectations of confectionery producers are changing at a rapid rate, with personalisation, customisation, and individualisation proving to be some of the most significant buzzwords in recent years.
Teamed with the wellness movement and the diversification of the global food chain, there is an increasing amount of pressure on manufacturers to turn around products at a shorter pace while maintaining a high standard of quality and safety. To handle these new requirements, the reliance on the use of automation, digitalisation and artificial intelligence (AI) within the food manufacturing environment has evolved from a need to a must in many cases.
An increasing number of manufacturers are adopting automation machinery such as robots in conjunction with technologies comprising vision systems or artificial intelligence (AI) to boost productivity and efficiency even further. The idea of AI in a production setting is nothing new, and is already being utilised by many manufacturers today. By essentially mimicking human activities using computer software, when added to the internet of things (IoT), it means that those devices can analyse data and make decisions and act on that data without the involvement of humans.
As outlined by Stuart Bashford, Digital Officer at Bühler Group, such systems as the company’s IoT solution Bühler Insights allows access to data and makes this available for visualisation in real-time by the customers. “We have used these advanced technologies (machine learning for example) on various projects, including in combination with our optical sorting technologies to help set up the machines,” explains Bashford. “Basically we can do supervised learning where we train the machine learning models using good and bad product, and then use this information to set the machine up for maximum efficiency.”
While there are many possibilities, Bashford highlights the importance of choosing the right technology to solve the problem. With such a diversified supply chain, the confectionery sector is wide-ranging and issues that need solving are unique to the supplier and will therefore require a unique technology. “Just because we have capabilities in artificial intelligence and data science does not mean that every problem can be solved with it,” adds Bashford. “There is so much potential from these advanced technologies that can be taken advantage of; the real trick is to focus on a problem, and then apply the appropriate technology to solve the issue.”
Converting data into actionable insights
Operating within an asset intensive industry means confectionery producers must meet quality demands, regulatory requirements and cost pressures, not to mention the growing focus on
sustainability and waste reduction. Ensuring equipment is operating at the highest levels of productivity and efficiency is therefore vital. However, the significant amount of failures are caused by incorrect operating behaviour, emphasises Marcello Gulinelli, Global Head of Food and Beverage at ABB Industrial Automation.
“To tackle these behavioural failures, it is vital to understand and work to the operating limits of the equipment and systems in place,” explains Gulinelli. Data capture technologies are integral to providing the information required to overcome such issues, and yet without analysis tools, there is no way for a manufacturer to derive actionable business insights and tangible benefits.
Advanced analytics platforms such as the recently released ABB Ability Genix can provide this link in the chain by contextualising operating (OT), information (IT) and engineering technology (ET) data to provide actionable insights that empower better decision making. For example, operators are able to see how equipment and systems are performing against different dimensions of asset performance such as quality, cost and safety. Quality managers can then forecast the quality deviation of a batch cycle and decide whether to stop the process and fine tune it.
A first step in the process is to collect the OT data, which Gulinelli notes poses unique challenges for the confectionery sector as varying machines may be running on different controllers. “ABB Ability Edgenius Operations Data Manager, a key component of ABB Ability Genix, is the link,” adds Gulinelli. “It is able to connect to, collect and analyse OT data at the point of production to improve processes and asset utilisation, while feeding the data to ABB Ability Genix, where it is combined with IT and ET data to bring strategic value to the business.”
Another important element is the scalability of the system, something that Gulinelli highlights is a key aspect of the ABB Ability Genix, which can be implemented on a flexible, modular approach. “For example, it could be implemented to optimise a single asset or value driver but can then easily be scaled up as needed,” he adds. When seeking data analytics software and services, it is therefore essential to seek a provider that can work with and for the unique circumstances of your production line to offer benefits no matter what stage of your digitalisation journey.
Reducing waste with smart software
In terms of waste reduction, automating the planning process can significantly slash the amount of food that is produced and not used. Supermarkets can make use of production planning and forecasting systems such as BakePlan, which allows in-store bakeries to predict what needs to be made each day.
Speaking about the issue of food waste, Séamus Quinn, Communications Manager, Cybake USA, says: “If planning is done incorrectly by guesswork or primitive calculations, which, sadly, is the norm, supermarkets and convenience stores make the wrong products at the wrong times. This, of course, contributes to eye-popping amounts of food waste.
“Even if you get your forecasting right, if you haven’t got a way to automate the organisation of your production waves throughout each trading day, staff end up making the right products in the right amounts but at the wrong times, which leads to food waste anyway.”
Therefore, the concept behind BakePlan is to prevent food waste being made in the first place while increasing sales and consumer satisfaction in a highly competitive sector. Similarly, Cybake, a bakery management software system, reduces waste by taking the guesswork out of batch size calculations. By switching to a cloud-only subscription model, the company appeals to a global industry, allowing store managers to change product orders and record waste and deliveries. This information is sent to users electronically, reducing manual processing and improving accuracy.
AI at the Edge
AI is already being implemented in many confectionery settings to aid in the analysis of large amounts of data to improve R&D, plant management processes and adaptability in packaging, as well as prolong equipment longevity and detect unforeseen events to prevent failures. Cloud computing has been established to process and evaluate the growing flow of data in production, with many of the AI solutions advertised on the market being Cloud-based. However, these solutions have significant requirements in terms of infrastructure and IT, and work with an overwhelming amount of data that can be laborious and time-consuming to prepare and process.
“The question of added value often remains somewhat murky for providers, who cannot determine whether or how investing in AI will provide a return,” explains Patricia Torres, Industry Marketing Manager Food and Commodities, Omron Europe. “The fact that system designs for the production industry are generally both complex and unique is another contributing factor.”
So while the Cloud is best suited to deal with big data and manage massive long-term analytics, AI at the Edge is crucial for real-time applications if a confectionery supplier is to integrate AI that creates tangible added value.
This approach offers more flexibility and faster response times, so production environments can get better use of data analysis at the Edge. “Instead of laboriously searching a huge volume of data for patterns, in addition to the processes that are running, it’s important to tackle things from the other direction,” explains Torres. “Technology is needed where the required AI algorithms are integrated into the machine control system, thus creating the framework for real-time optimisation truly on the Edge – at the machine, for the machine.”
One good example of this technology is Omron’s Sysmac AI Controller, a smart AI solution that collects, analyses and utilises data on Edge devices within a controller to prolong equipment longevity and detect abnormalities to prevent failures. It combines control functions of manufacturing lines and equipment with AI processing at manufacturing sites in real time.
Moving to the Edge
Another example is Siemens Industrial Edge, a digitalisation solution that adds machine-level data processing to automation devices by taking the intelligence of Edge computing and thus sophisticated analytics securely to manufacturing level. Cloud connectivity is used in conjunction with Edge applications in an integrated hardware and software ecosystem for automation components. As outlined by the company, storage and transmission costs are reduced for users because large volumes of data are preprocessed, and only relevant data is then transmitted to a Cloud or IT infrastructure.
But Siemens signified that Edge computing is not an end in itself, but a means to achieve specific goals based on the unique needs of the manufacturer. Cloud and Edge computing aren’t mutually exclusive, but are conditional, and when deciding on one of the approaches or a hybrid solution, it is important to consider the framework conditions and the business objectives of the deployment. This is particularly true for food and beverage firms including confectionery, where production facilities are often outdated and investment funds are low. Find a technology provider that can provide a step-by-step approach to implementing Edge computing and how it can benefit your business.
Finance options for smart factory solutions
Of course, the same rule applies to AI at the Edge as with all Industry 4.0 technologies – for a confectionery manufacturer, it is vital to consider both the provider of such solutions and the technology that is best suited to your own unique challenges and requirements. “Automation in this context does not only mean robotics or artificial intelligence, but a well-thought-out overall structure of fixed, collaborative and mobile robotics; plus monitoring and control technology, sensors and vision technology tailored to the respective production requirements,” explains Robert Brooks, European Marketing Manager, Food & Beverage, Omron Industrial Automation Europe. In addition, the various stakeholders and market drivers should not be considered in isolation, but as a whole and integrated into the future production strategy.
However, many confectionery companies might be reluctant to invest in additional technologies, particularly following the economic difficulties presented by the coronavirus pandemic. In this instance, cost structures with leasing or finance options are being implemented by companies including those discussed as an option to allow adaptability on the factory floor. This way, it’s possible to continue to maintain a competitive edge and increase your return on investment, without putting a strain on existing capital or credit lines.