The LDF initiative and network is an excellent example for all strategic and reputational aspects of Siemens. By making use of Siemens own Hardware and Software solutions, the participating plants make valuable experiences, which often also create helpful product management feedbacks. By this, it’s a great support of “We Use What We Sell” and “We Sell What We Use”, both supporting the “Technology with purpose” aspect. Through these means, our Siemens plants became even more resilient, what helps our customers to still get their ordered innovative products from us, so “Customer impact” is highly positive. All achievements are based on great “People Empowerment” approaches paired with modern agile working models. The different skill development offers support the “Growth Mindset” of our colleagues. By using LDF principles and results now Siemens wide through the Factory Digitalization program, the experiences and (newly) created scaling technologies are used in the different ecosystems of plants and colleagues, all in the sense of One Siemens. By acting not only internally, but also externally by making use of the reference and show cases for digitalization, the created solutions across the different BUs and plants stimulate our sales, revenue and profit, to make Siemens even more sustainable successful. By collecting different external awards for being a digital thought leader and Industrie 4.0 driver, LDF strengthens Siemens’ reputation and as a consequence, should be also awarded internally with the “Werner von Siemens Award 2021” … !Klaus Oesterschulze, Global Head of Information Technology DI IT
Workstream “Digital Twin“
Let us use the workstream digital twin as an example. On the left side you can see the main attributes this workstream addresses. In this case, it addresses speed and efficiency (three squares) mainly and flexibility, quality, and scalability and sustainability (one square) to a certain degree. In the box below, you also see, which solutions we use out of the Siemens portfolio to reach our goals.
Verbally, this roadmap shows our mission, that production engineering is automated, enabled by full-featured Digital Twins of Product, Production & Performance. These digital twins are aligned to the ECLASS cross-industry standard and contain all levels of detail to automize the process from product design to production on the shop floor. The high degree of automation based on artificial intelligence (AI) and supported by a knowledge database raises the efficiency in engineering and speeds up time to market to lay the foundation for shorter innovation cycles.
This is supported by the automated programming of machines (in Mentor Valor PP or NX CAM). All tasks are coordinated by an overarching Mendix workflow system, guiding through the whole process and ensuring fast execution. Seamless interoperability between all involved systems eliminates manual efforts for data preparation and raises output efficiency and quality of production engineering by providing comprehensive data.
Integrated simulation of the production process in the virtual factory increases the reliability of production engineering. This ensures that the engineered production system will deliver the targeted performance with zero defects in the first place. Therefore, all tasks from work cell design to optimization in the context of the whole factory infrastructure including logistic, workforce and maintenance are digitally assisted with Process Simulate, Line Designer and Plant Simulation. Automatic human work design calculation is carried out (TiCon4Teamcenter), while generating new work plans for process optimization.
The digital twin of performance closes the loop to the real-world production by providing real-time data from the Manufacturing Data Ecosystem to initiate optimizations according to actual boundary conditions with AI. The workstream digital twin is also described with so called reference processes.
One section of the reference process called “production system planning”. We defined the input and how the information should flow through the several process steps. It is defined which software is used and what is automated, semi-automated and still manual.
As an example, we are aiming to have a rough plan of our production equipment in just 6 hours after we received the design information of our product via a digital twin of the product. Due to the digital twin of our production, we can convert the information from the product engineering department in extremely short time.
For example, in a certain step of the reference process „New product Introduction” and „New Machine Introduction” our engineers had to order stencils for soldering paste printing. This had to be done several hundred and hundreds of times in our factories manually.
With a workflow programmed with Mendix, a low code application by Siemens, we simply automated this task, and no manual ordering is required any more. This is just one example of more than 30 sub-processes which were automated.
The Workstream “Processes”
In our next workstream, “Processes” we address an autonomous End-toEnd-coordination of supply chain resources based on artificial intelligence and real time transparency to maximize speed and efficiency.
All resources (material, machines, workers, tools, and fixtures) along the end-to-end supply chain are simultaneously considered in both capacity utilization and prediction (short & long-term) to maximize production efficiency of machinery and workers, using Opcenter APS.
Products and production can directly communicate with each other for decentralized, autonomous self-optimization of production execution under consideration of actual boundary conditions to reduce manual effort for production planning and control on shop floor level. Therefore, an agent-based architecture of cyber-physical production systems is the blueprint to raise the flexibility of our Manufacturing Execution Systems to the next level with adaptable and distributed microservices.
In the third workstream line of “Processes”, we describe that real-time data from the supply chain are monitored continuously to detect deviations at the earliest moment to increase the options for action. Therefore, external data from the Supply Chain Suite are considered simultaneously with internal material and resource position information from the Real Time Location System (SIMATIC RTLS or 5G) with the accuracy of just centimeters. Based on this transparency the system automatically provides valued solutions in case of deviations with a self-learning system to increase the efficiency in deviation management.
Also, this workstream is based on reference processes for example the inbound and intralogistics process. The simple vision we are sharing here is, no touch order execution. If the demand is coming in from the costumer, automatically it will be checked if the required resources of people, machines and material are available and if so, a smart algorithm defines the production line, order sequence and will release it to production.
Another visionary tool we developed is the so called “Dynamic Value Stream Analysis (DVSA)”. In most of our electronic factories, a value stream for a product is not always linear and thus very complex due to machine and resource sharing. In addition, decentralized full automated material handling hides partially material stock and high product variance with demand variability makes it even more complicated to optimize value streams in our factories with regard to through put time, inventory and OEE. The DVSA leverages already existing shop floor data, so no additional sources are needed for turning big into smart data. Based on the raw data, several production insights can be generated purely by applying mathematical methods and cloud resources are being leveraged for real time calculation of value stream KPIs & data visualization. The transparency from DVSA enables us in many ways to improve our production and engineering processes. By having instant transparency at hand, we don’t have to waste time with collecting data – we can start with problem solving right away. New information like change over matrices, that previously weren’t available due to high variance, is being uncovered and utilized for minimization of downtime