Process Mining is the (r)evolutionary progression from value stream analysis
in PAFnow | Aprendiendo

Process Mining is the (r)evolutionary progression from value stream analysis

The classic value stream analysis has been a popular method for years to uncover optimization potentials in companies by identifying non-value-adding processes and checking operational processes for their usefulness. However, an increasingly fast working world and digitalization are pushing analog and long-term value stream optimization projects to their limits. Although companies already have access to faster, digital solutions, they often shy away from using them.

Concerns about data protection, fears over jobs and, in some cases, misunderstandings about new technologies complicate the transition to digital solutions.

Yet a study by the Fraunhofer Institute for Industrial Engineering (IAO) shows that by far not all problems are found and solved with the value stream method. Optimization potentials are left undiscovered. In addition, traditional value stream analysis is completely unsuitable for flexible or complex processes and structures - the greater the number of process types and variants, the less effective the value stream method.

In addition, the analogous workshop structure of the value stream analysis binds several employees in an optimization project in the long term, can interfere with operational procedures and is only possible on-site.

All these weak points require a solution.

Process Mining offers this solution as a (r)evolutionary progression from value stream analysis. (Learn more in our PAFnow OPEX Edition webinar recording)

In simple terms, Process Mining follows a data-based approach that reduces the collection and mapping of information to just a few days, but still makes problem solving in a team interactive. The philosophies and ideas of classical value stream mapping thus remain, but are simplified, modernized, and accelerated.

From Value Stream Analysis to Value Stream Analysis 4.0 to Process Mining

Classic value stream method

The traditional value stream method consists of three steps

  • Defining added value
    First, an awareness of customer benefits as well as waste must be established and the possible areas of application and prerequisites for an analysis must be examined.
  • Analysis of as-is situation (value stream analysis)
    Analogue recording of the process through workshops, interviews and on-site inspection by a project team. After recording all necessary data and times, the value stream flow is manually documented in a suitable diagram and wastes are identified.
  • Defining the target model (value stream design)
    A new value stream is defined, and waste is eliminated.

Value stream analysis 4.0

While the classical value stream analysis concentrates on material flows and associated control information in the production and service sector, the value stream analysis 4.0 extends the examination to information flows. By including the information flow, a holistic view of the underlying processes is obtained.

The systematic consideration and cataloging of information flows adds new steps to the value stream method.

VSA 4.0 extra steps
Value stream method 4.0, Source: 'Abbildung 15' Leitfaden Industrie 4.0 trifft LEAN, VDMA[1]

The conventional representation of the value stream is extended in such a way that information streams between the individual sources can be captured and displayed:

vsa 4.0 schema
SCHEME VSA 4.0, Source: Institut für Produktionsmanagement, Technologie und Werkzeugmaschinen (PTW), TU Darmstadt[2][3][4]

“This opens up a new perspective on the provision and sharing of information with regard to

  • Wastefulness in dealing with information
  • Use of information for process improvement
  • Use of information to increase customer benefit

Value Stream Analysis 4.0 is thus more comprehensive and holistic than the classic method, but problems such as the time factor, the project structure and the analog and face-to-face character remain."[2]

Process Mining

Process Mining can overcome the disadvantages of value stream analysis, provided the technical requirements are met. It can be seen as a digital supplement to the value stream method.

Instead of workshops, an on-site inspection and employee interviews, Process Mining relies on the process data already available in the system. Nowadays, these are saved in every company’s IT system, but still not used sufficiently.

This often has two reasons. First, companies are not aware that they can turn their data into process knowledge. Secondly, they often don’t know what data is needed and how it can be accessed. Here, Value Stream Analysis 4.0 provides clarity about the current status and identifies gaps that need to be filled in order to achieve the target status in the company.

Under the appropriate technical conditions, data is always available as an excellent source of information and is less prone to errors than the analogous collection of information.

In addition, they are collected continuously, practically automatically, and can be constantly updated.

Process Mining software uses this data to automatically create an image of the process, including all variants. The process graph obtained in this way contains all the information from millions of data points. In contrast to value stream analysis, it also records complex processes with their diverse, often invisible variants and can be analyzed down to the smallest detail on an individual level. Furthermore, intelligent Process Mining software recognizes and marks longer processing times and violations in the process flow.

Process Mining visuals
The Process Mining solution PAFnow of the Process Analytics Factory offers various views on the process, here the characteristic process flow and the swimlane view of the PAFnow Case Viewer, which shows organizational connections.

Process Mining tools, which are fully integrated into Business Intelligence solutions, also enable a detailed analysis and easy-to-understand visualization of process flows. One such tool is PAFnow, which embeds Process Mining directly into the Microsoft 365 environment through integration with Power BI, the market leading solution. This opens up completely new possibilities for direct process optimization.

Via Power BI, Process Mining users have access to over 150 connectors to data sources and over 300 action connectors to programs with which actions can be initiated directly, such as collaboration, project management, workflow automation, RPA and the connection of flows and apps with leading applications such as SAP® ERP (ECC and S4/HANA), Salesforce or ServiceNow.

The results of the analysis can be easily and effectively shared with the relevant employees. It is also possible to access the analyses from mobile devices, i.e. when on the road, and to react flexibly to problems. The program supports this with data alerts and automation options for simple work steps.

This results in a new methodology that delivers the first results within a few weeks, constantly reacts to current information, and enables targeted optimization by responsible employees.

the three phases of Process Mining projects
Three phases of the Process Mining method - analysis - joint actions - automation.

So with Process Mining there is an evolution from static to dynamic analysis and a revolution from analog to digital working.

Advantages from a corporate strategy perspective

If the value stream method is combined with Process Mining, there are many advantages and opportunities for companies, so that digitization no longer represents a challenge but an opportunity.

Basically, with Process Mining it is very easy to use data in such a way that customer-relevant information can be found. But this only works if the data can actually be used. Here, the value stream method provides information about storage media, collected data and possible gaps that must be closed before Process Mining can be used.

Once the path is clear for Process Mining, new digital business models can be developed with the information gained, while existing processes are continuously improved. In this way, data does not end up as “digital wastes”, but actively contributes to the success of the company.

There are also advantages with regard to digitization strategies. On the one hand, Process Mining avoids the “blind” digitization of broken processes and procedures, since it provides full transparency of all process variants. On the other hand, Process Mining supports all three phases of migration or transformation to cloud-based ERP systems. As a result, errors are minimized while ensuring that normal operations can continue during the transformation.

In addition, there are “classic” opportunities for optimization, such as the reduction of lead times, the flexible alignment of supply chains and the early detection of bottlenecks.

Plus, the combination of Process Mining and Business Intelligence enables the solution of problems in the information flow, brought to light by Value Stream Analysis 4.0. At the same time, weaknesses in the value stream analysis are compensated.

It is very easy to build information streams within the Microsoft 365 environment, store data in a central location and manage access rights. In this way, information flows can be optimized directly and, if not already done, the prerequisites for successful Process Mining can be created.

The subsequent analysis with Process Mining runs in the background, so that the operative daily business is not interrupted. While the first optimization measures are already underway, the analysis can continue continuously. In this way, one also gains timely insights into how effective the measures taken actually are.

Thus, Process Mining enables the development away from process optimization projects and towards continuous improvement.

Perhaps one of the biggest advantages of Process Mining has become clear in the last few months. Due to the global pandemic caused by Covid19, many companies had to struggle with delivery shortages and liquidity problems, while at the same time almost all employees had to work from their home offices.

Process Mining could still be used in these circumstances so that companies were able to react quickly to the new conditions and problems. Both in supply chain management and in the ability to deliver to customers, lead times could be kept short and inventories could be managed flexibly. As a result, companies using Process Mining were able to keep their production running at the best possible level despite global bottlenecks.

The remote character of Process Mining clearly sets it apart from the value stream method and has proven to be a valuable advantage during the crisis. In addition, many companies were able to experience the advantages of Microsoft 365 directly. The structures to start successfully with Process Mining in Power BI were created in the last months at the latest.

Best Practices – The Example of Hottinger Brüel & Kjaer

Process Mining is worthwhile wherever processes take place, regardless of industry or company size. To achieve the desired success with Process Mining, however, you need a strategy that takes both technical and organizational requirements into account.

Hottinger Brüel & Kjær (HBK) uses a combination of Value Stream Analysis 4.0 and Process Mining. The company is very successful with it. Karl-Heinz Pöhlmann, Vice President Supply Chain, explains: " The development and introduction of Value Stream Analysis 4.0 was already an important milestone in the further advancement of our Lean Management. Process Mining is the logical next step”. HBK’s strategy follows a top-down structure and involves employees in the decision-making process. The basic procedure follows a clear scheme:

hbk best practices
Value stream method 4.0: HBK's strategy.

For the introduction of Process Mining, HBK has defined the ambitious goal of increasing its delivery capability in the order-to-cash process to 95 percent. This goal was achieved in significantly less time than with a traditional approach.

For a smooth start, HBK recommends a three-step approach from Value Stream Analysis ( drawing) to Value Stream Analysis 4.0 (include data, make data usable) to Process Mining (further process data, optimization) and a clear goal for the introduction of Process Mining. The procedure is based on the structure of the value stream strategy.

First, the management, works council and other decision makers from the individual departments are informed, an overall goal is formulated and consensus is built. It is important to address the concerns of the employees. It is not a matter of monitoring or exposing them. Data is anonymized and made available in a targeted manner; it is not used to measure performance, but to optimize processes. The question is still not “Who?” but “Why?”.

Then it is a matter of clarifying the technical requirements. Where does the required data come from? Is the existing data usable? Which software is available?

After that, suppliers are sighted and proof of concepts are carried out. Once the decision has been made, the evaluation begins with Process Mining.

HBK has decided in favor of PAFnow, because the connection through PAFnow with Microsoft 365 allows the possibilities to go far beyond pure Process Mining and processes can be continuously optimized from within the program. Here it becomes clear once again that Process Mining is a continuous solution and not an isolated optimization project.

Outlook

The fields of application for Process Mining are virtually unlimited, since today data on processes are available everywhere, even if they are not actively and consciously collected. Typical areas of application are found in production/manufacturing, purchasing, healthcare and the financial sector, but also e.g. for customer journeys, digitization, migration and transformation projects, the digital workplace, auditing and in process management.

Due to the broad availability of data and an increasingly dense and far-reaching networking within companies, new possibilities to combine Process Mining with other technologies are constantly arising in order to exploit optimization potentials to the fullest possible extent.

Already today, Artificial Intelligence and automation play an increasingly important role in the optimization process.
The incredible amount of data created every day is beyond the reach of humans alone. This is where Artificial Intelligence helps to evaluate the data. AI-supported analysis makes it easier and faster to get started with Process Mining and find problems that would otherwise remain hidden.

Even the often ‘demonized’ automation aims at support. It is intended to cover monotonous and repetitive tasks, so that employees have more time to concentrate on complex and intricate issues.

Thus, the combination of value stream method, Process Mining, and automation results in value and information streams that can flow continuously, contribute to continuous process improvement and support people in their work.


Sources mentioned (used with permission)

1 Leitfaden Industrie 4.0 trifft Lean Wertschöpfung ganzheitlich steigern, Hartmut Rauen, Dr. Christian Mosch, Felix Prumbohm, VDMA
2 Value stream mapping 4.0: Holistic examination of value stream and information logistics in production, Tobias Meudt, Joachim Metternich, Eberhard Abele, 2017, CIRP Annals
3 Wertstromanalyse 4.0, Tobias Meudt, Markus P Rößler, Jörg Böllhoff, Joachim Metternich, 2016, ZWF Zeitschrift für wirtschaftlichen Fabrikbetrieb
4 Value stream method 4.0: Holistic method to analyse and design value streams in the digital age, Lukas Hartmann, Tobias Meudt, Stefan Seifermann, Joachim Metternich, 2018, Procedia CIRP


Authors

Karl-Heinz Pöhlmann is the Vice President Supply Chain at Hottinger Brüel & Kjaer GmbH (HBK) and responsible for purchasing, production, logistics and technology. He has been working in the automotive and high-tech sectors for over 20 years and has been deeply involved in value stream mapping and data analysis in the production environment for many years. HBK uses a combination of WSA, WSA 4.0 and Process Mining to optimize its processes. PAF and HBK have been working together for several years.

Katharina Laumann is Content Manager and Editor at Process Analytics Factory (PAF). She is a professional journalist who specialized in the fields of natural sciences, technology, economics, and research during her studies.