Filling technology

Mr Vanhöfen, to what extent can AI-supported analyses contribute to improving sustainability in food production?

AI enables resource-saving planning by manufacturing production quantities according to demand and minimising the use of resources. For example, this reduces spoilage due to overproduction. This is because modern forecasting algorithms can predict requirements more precisely, leading to a reduction in stock levels, greater delivery reliability and ultimately the minimisation of supply chain risks. Energy and material consumption can also be optimised. When using renewable energies - such as solar power - AI can control consumption to minimise peak loads and make efficient use of energy availability.

What is the focus of OMMM's activities?

With our AI-supported approach, we optimise industrial planning, especially in the process industry. Our software takes the entire supply chain into account. This means all stocks - from primary materials to finished products - and includes them in the planning. In combination with precise demand forecasting, this leads to significant inventory reductions, lower storage costs and better capital commitment - resulting in clear efficiency gains.

How high do you rate the need for these kinds of solutions?

Many companies often still work with analogue methods for industrial planning. These are usually not transparent and take hours or even days to achieve a planned result - leading to unnecessary personnel, set-up and cleaning costs as well as high reject rates. However, if food producers want to make their supply chains resilient, sustainable and flexible, this can only be achieved through a higher degree of digitalisation and the use of advanced AI.

How can the success of an AI implementation in supply chain management be measured and what key performance indicators do you use?

Success can be measured using monetary KPIs such as set-up and cleaning costs, storage costs, capital commitment, costs of delayed production and costs for responding to disruptions. Other relevant indicators are plant efficiency, degree of digitalisation, sustainability, responsiveness and customer satisfaction.

Can you give specific examples of how food manufacturers have achieved significant savings by implementing your planning solutions?

Zentis now uses our software in all its plants in Germany, the USA and Poland. Thanks to optimised planning, the company saves over 1.2 million euros per plant every year. The set-up and cleaning costs were reduced by 30 percent or 750,000 euros and the costs for the disposal of rejects by 50 percent or 500,000 euros per year. As fewer employees are needed in planning, the staff now have free capacity for other projects within the company. Other users include Schluckwerder and the Valeo Foods Group.

How do you ensure that your software fulfils the complex requirements of the food industry? Especially as this is a very heterogeneous industry ...

In fact, every production setup is unique and presents a number of special challenges. Our AI is able to digitally mirror the existing setup down to the smallest detail to enable an unprecedented level of planning automation and optimisation. Our co-founder, POM Prof. Tempelmeier GmbH, has created the conditions for this and has conducted pioneering work in the entire field of capacitive production planning. In addition, with our shareholder POM, we at OMMM have more than 20 years of experience in implementing customised planning solutions in a wide range of industries. We understand and know the daily challenges of the shop floor - a multitude of complex rules, exceptions and restrictions.

What advice would you give to small and medium-sized food companies that are looking at AI-based SCM solutions for the first time?

In principle, manufacturers need to know exactly where they stand. That's why we often carry out a digital readiness assessment first. Here, we focus on all relevant processes along the entire value chain and assess the degree of digitalisation: If this is low, it's one. If it's high, however, it's four. This is followed by an analysis of the weak points and the most significant cost drivers along the supply chain.

... analysing data therefore plays a major role in your approach. What data is included?

Anyone involved in production planning optimisation in the food industry should ensure that high-quality forecast data is available in addition to the order data. If this is not the case, we recommend starting with AI-supported forecasting systems. At the same time, the data quality in existing ERP, MIS and warehouse management systems should be consistently reviewed and improved where necessary.

Which sources are used and how can data quality be ensured?

The quality of existing data in companies is often one of the greatest hurdles. Nevertheless, consistent data integration from different sources is a decisive success factor. Systems such as ERP, MIS/BDE or external warehouse management systems must also be connected for a holistic solution. The quality of this data must be optimised through structured measures where necessary. Added to this is the integration of previously unused but potentially valuable information - for example, transport logistics capacity limits between production stages. This type of data has often not been relevant or has not been recorded in the past, but must be available and systematically mapped for AI applications.

How does the OMMM solution integrate into existing IT infrastructures, for example ERP systems like SAP?

OMMM can be seamlessly integrated into existing system landscapes by importing and exporting the required data. We have official certification for SAP so that we can seamlessly embed our planning solution in SAP's process and data landscape and replace its generic planning module. Data is usually exchanged via iDocs. This integration enables smooth operation without media disruptions.

What future developments and enhancements are planned for OMMM's solutions?

We are currently piloting both detailed personnel scheduling and a complete solution for materials management. We offer these as add-ons to the existing Demand, Production and Business Planning and Supply Chain Dashboard modules, which will soon be available as standard modules. We are also developing AI-supported assistants to support daily planning tasks and accelerated implementation methods based on artificial intelligence. Overall, we will significantly increase our level of standardisation and provide ready-to-go solutions for different target groups and sectors in the process industry, such as chemicals, pharmaceuticals and flow finishers in general.

For additional information, go to:

OMMM Operations Management Solutions GmbH
Leverkusen, Germany
Reinhard Vanhöfen
vanhoefen@ommm.ai
https://ommm.ai/