Globalization is creating a turmoil in the manufacturing industry and existing ERP and MES systems are not offering the flexibility that is needed when suddenly competences are missing due a COVID outbreak or suppliers are simply unable to deliver material in time. Dealing with this scenario requires pragmatism and creativity. Unfortunately existing Production IT systems are not designed for creativity, they are designed for long term planning.
Trying to define process adaptations in systems such as SAP ME is a tremendous challenge and typically, out of desperation, production engineering turns to universal tools such as Microsoft Excel. The result is hugely complex XLS files that are hard to maintain. Additionally all the operational data is lost and it becomes even harder to perform any kind of analysis and optimizations. What is even worse than this is that these kind of “tools” typically exist for a very long time and become part of the shadow IT that sooner or later becomes unmanageable. Latest when the creators leave for another job.
We have built the WORKERBASE platform to overcome the severe limitations of legacy production IT systems. Even before the pandemic and the war in Ukraine, it became evident that there are an ever increasing number of disruptions in production flows, that cannot be prevented by or handled in traditional IT systems. What is required is a realtime view of the production including all aspects: man, machine and material. Creating such realtime data is not sufficient though, since it is also required to be able to make realtime decisions. Quite often existing systems are used to produce management dashboards that try to answer the question: “What went wrong yesterday?”. Management then deals with the problem of trying to identify the root causes and define mitigating actions.
What if you would not need to ask that question anymore? What if you had a system that produces such dashboard data and at the same time automatically runs the mitigating actions combining system intelligence with human intelligence?
This is WORKERBASE and this is what we call Dynamic Process Execution.
To cope with the requirements of an ever-changing world, modern production systems need to fulfill new requirements:
By using digital workflows to connect their frontline workers with machines and IT, manufacturers ensure that all relevant data sources are connected and used to create an accurate real-time picture of the whole production. Datasources include human workers, machinery and IT systems.
By dissecting large workflows into small activities, manufacturers can ensure that each activity is dynamically executed. This needs a radically different IT architecture, compared to traditional MES systems. Modern systems such as the WORKERBASE platform follow an event driven architecture and allow to build a more distributed and real-time production system that dynamically orchestrates distinct activities.
The WORKERBASE system builds up several layers of context that can then be used to make system driven decisions that – as opposed to human decisions – deliver optimal outcomes by taking into account all data and all options. We directly execute these decisions in the form of production processes performed by human workers.
Chunks of work processes are defined in independent layers that are self optimizing against each other based on a set of common business rules (e.g. logistics only needs to act when production has finished,…). The self optimization is performed by system calculated decisions combined with human decision making and dynamic workload allocation. The system knows the skills and status of each worker and during process design these parameters are used for designing workflows based on simple “cause and effect” logic which describes which worker should perform which task at any given time. The system can therefore dynamically determine the best location for a task, calculate correct timing when a task should execute, change the sequence of worksteps or products and forecast workload. Using the system’s universal user interface, workers are able to pick up any device to get guidance and interface with the system functions.
A modern production facility will be completely data driven. All data from all sources, including human workers, is always available in realtime and used to dynamically steer the production from machines to material to humans.
By using real-time data to generate forecasts, systems will be able to adjust the production flow at any time to ensure an always optimal utilization of all resources. These systems are connected between manufacturers and their suppliers driving an optimal flow from the end to the start. These systems allow a new way of production that is availability driven and less likely to be affected by supply chain and quality problems. Using forecasts in all areas from predictive maintenance to prediction of workload allows to remove close to all delays due to unplanned interruptions.
Systems create configurations of material, machines and workers that allow a configurable range of products to be produced at any time. Once all components are available, these systems then schedule a dynamic production flow that takes into account all parameters from planned delivery date to worker skill availability. Large portions of today’s production complexity will be removed by simple business rules that automatically execute without human intervention (e.g. material replenishment and ordering process). A range of production related jobs that are not in the blue collar domain will disappear such as worker coordination, purchasing/dispatching and production planning in the classical sense.
The pressure on the efficiency of manufacturing has increased tremendously. We are living in a VUCA world characterized by high volatility, uncertainty, complexity and ambiguity. Manufacturers are facing many challenges with high energy costs, volatile customer demand, missing materials and Customers pushing for shorter fulfillment times. In the past years, companies have focused on creating stable and lean production processes. But the LEAN methodology does not provide enough benefits anymore, the new reality calls for new approaches on how to organize production. To guarantee stable operations, manufacturing plants rely on statically planned work procedures. But with volatile markets, uncertain supply chains and complex customer requirements these procedures are frequently interrupted and the planned state never manifests. What’s even worse: current production systems such as MES are not prepared to cope with the new demand. As traditional production systems were built with a focus on providing stability, they typically use static planning and are not able to cope with the new demand.
Let’s look at one example: to circumvent unplanned disruptions in the area of material provisioning, buffers are placed in the production flow in order to accommodate disruptions. This unnecessarily binds capital and the buffers are never right for the situation, they are either too big or too small. Due to not having a full real-time view of the whole production status, it is impossible for manufacturers to optimize their flow for the actual situation. This severely impacts operational performance by
If production can be streamlined with real-time data and real-time decision making, production cycles will shorten, work in progress can be reduced and buffers regarding time and material can be decreased. This creates more output with the same layout and staff and reduces production cost drastically. This is exactly were Workerbase comes into play. WORKERBASE combines human workers, machines and material into one digital Dynamc Process Execution platform that enables real-time and data-driven production processes. Production planners can plan their processes in a much simpler way since they do not need to look at the End to End chain of tasks. Instead they can break down the whole process into small chunks that only concern a subset of the people, for example coordinating forklift drivers. This allows to construct an optimal process in an encapsulated way where the expertise of people performing such jobs is directly transferred into the digital workflow. The result of this dynamic workflow modeling is that areas of concern are loosely coupled and self optimizing in a way that if a task is created in a new area, the system will determine the optimal way of execution (e.g. directly send a notification to a forklift driver). The individual workers are then given some degree of autonomy to optimize their own work processes. The result is an optimal combination of system and human intelligence.