AI-as-a-Service

Intelligent production planning in seconds instead of hours

AI planning made easy

Our solutions can be seamlessly integrated into your existing MES, PPS, or APS software. They deliver optimised and traceable planning results – without the need for in-house AI development. AI becomes a 24/7 sparring partner that provides continuous support and relief.

Making complexity manageable

Many combinatorial problems in production planning are too complex for traditional methods. With our approaches, even highly complex scenarios can be solved efficiently – automatically and in the shortest possible time.

Numbers instead of gut feeling

Mathematical optimisation delivers reliable decisions: greater adherence to deadlines, shorter set-up times, optimal warehouse utilisation, and more. This not only leads to faster results, but also to demonstrably better ones.

Secure a free consultation

Our solutions in action

We solve typical production planning challenges, from detailed planning and resource allocation to production levelling. Our algorithms can not only completely recreate production plans, but also specifically improve existing plans that have been created manually.

Detailed Planning

Optimised detailed planning for complex set-up processes

This optimisation scenario addresses detailed planning in discrete manufacturing with multiple parallel production lines. Set-up times occur between orders, which can vary depending on the machine – as can processing times. The algorithm also takes into account shift calendars, secondary resources and complex dependencies between orders in order to create realistic and efficient production plans.

Automatic planning for the food and beverage industry

This optimisation algorithm was developed specifically for typical scenarios in food and beverage production. It takes into account both the manufacture of intermediate products and their packaging – including limited capacities in intermediate storage facilities. Individual restrictions and priorities can be flexibly mapped, such as which products may be stored where or should be given preference.

Intelligent production planning for the process industry

This algorithm automatically creates optimised production plans for the process industry – for example, in mixed feed production. It takes into account factors such as contamination risks, time intervals between orders, incompatible sequences and the availability of equipment and intermediate storage facilities. The aim is to minimise delays, set-up costs and cleaning, ensure even utilisation of resources and avoid downtime.

Efficient planning for paint shops

Planning paint shops is complex – numerous parts have to be efficiently controlled through the plant every day. Goods carriers move via conveyor systems, and planning is done in rounds. Factors such as colour sequences, block sizes, cleaning processes, technical restrictions and parallel booths increase the complexity. The aim is to create a plan with as few colour changes and low costs as possible – taking into account deadlines and technical specifications. Our solution is suitable for painting systems in the automotive industry, the electronics industry and metal processing, among others.

Optimised round-the-clock planning for gear production

In circular systems for tooth production, the base material is injected into metal moulds and processed into raw teeth in a multi-stage cycle. The challenge lies in efficiently loading the system, taking into account delivery dates and the limited availability of the moulds as production aids. Similar planning requirements can also be found in rotary systems in the food, cosmetics, pharmaceutical, chemical and packaging industries.

Resource Allocation

Automatic allocation of limited resources such as tools

A production plan can only be implemented if all the necessary resources are available – such as production aids or interim storage capacities. The allocation of these secondary resources, such as boilers, tanks or tools, is highly complex and almost impossible to manage manually. Mathematical optimisation can be used to find solutions in seconds that are significantly better than any manual planning – fast, precise and resource-efficient.

Intelligent staff allocation

This algorithm assigns available employees to the appropriate workstations. Various objectives, such as employee or workstation priorities, qualifications or frequency of deployment, can be flexibly weighted and combined. The algorithm is called up from MCP Workforce Management, among other places, and ensures that planned orders are processed in the best possible way within a given time period.

From rough to detailed planning

Smoothing of production volumes according to the Heijunka principle

Production levelling involves distributing production volume (total and per product) evenly across individual periods. A period can be, for example, a shift, a day, a week or a month. The aim is to achieve an even utilisation of production capacity and a high degree of flexibility in response to fluctuations in demand. Production levelling is an important part of implementing the Heijunka principle. However, levelling production also often plays a decisive role in long-term capacity/production planning.

Formation of compatible batches for complex requirements

When planning orders in batches – for example, for heat treatment in furnaces – compatible orders can be processed together. Restrictions such as availability, release dates, set-up times and capacities must be taken into account. The aim is to create optimal batches to minimise running and throughput times – while meeting the required deadlines. Typical areas of application are heat treatments in electronics manufacturing and in the chemical, metal and packaging industries.

Are you interested in one of our optimisation solutions? Or would you like to tackle a different planning issue?

In a joint discussion, we will find out what makes sense for you.

Easy integration

AI …

Production planning is full of complex optimisation problems. It requires solutions that are intelligent, efficient and robust.

  • Mathematical optimisation: Methods such as constraint programming and integer programming deliver exact, traceable solutions – ideal for structured planning problems with clear constraints.
  • Metaheuristics: Methods such as simulated annealing efficiently traverse large search spaces and offer a flexible alternative when classical methods reach their limits.
  • Machine learning: Algorithms that learn from data, adapt and dynamically make better decisions. Complements mathematical optimisation methods with intelligent, data-driven support.

… as-a-Service

Our solutions can be easily integrated into your planning module – all you need to do is set up the interface. No complex project, no extensive adjustments.

  • Input: Your planning data – e.g. orders, resources, set-up times – is transferred to our AI via a REST API. The call can be integrated directly into your existing software.
  • Optimisation: The AI processes the data and calculates an optimised solution – quickly, reliably and transparently. Complex interdependencies are automatically taken into account.
  • Output: The result is an optimised production plan that is returned via the API – ready for display, further processing or transfer to downstream systems.

Extensive expertise

Scientifically based

Through regular exchanges with scientific partners, the latest research findings are incorporated directly into our development work – resulting in solutions that are technologically advanced and scientifically sound.

TU Wien

From 2017 to early 2025, our Christian Doppler Laboratory was based at the Institute for Logic and Computation at TU Vienna. Together with Bosch and Ximes, we conducted basic research there on new algorithms and the use of AI in production planning.

Since 2024, we have been funding a doctoral position at the Doctoral College iCAIML – thus remaining closely connected to research in areas such as hyperheuristics and automated algorithm selection.

KIT

Together with the Karlsruhe Institute of Technology (KIT), we are researching methods for making industrial processes more energy-flexible. The aim is to develop AI-supported methods that make energy-intensive production more grid-friendly and sustainable. This collaboration combines our many years of expertise in APS with KIT’s cutting-edge research in the field of energy system design. As an implementation partner, we translate research results into industrial practice – for sustainable production.

Uni Duisburg-Essen
Excerpt from research areas
  • Stochastic modelling
  • Capacity planning
Uni Siegen
Excerpt from research areas
  • Human factors
  • Industry 4.0
  • Decision analysis
  • Project management

Tried and tested

Thanks to our many years of experience from numerous industrial projects, we know exactly what matters in production planning.

We are familiar with the specific requirements and typical challenges of many industries, from mixed feed production to the food industry, from cosmetics and pharmaceuticals to electronics manufacturing.

Our solutions are proven, robust and flexible enough to deliver real added value in complex production environments as well as in software products for these industries.

Successful applications

Book an expert consultation