Process Mining as a Service

your first step towards implementing process mining
What is process mining?
Process Mining is a bridge between Data Science and Business Process Management. It applies Machine Learning and Data Mining techniques to comprehensive analysis of process event data. The goal is a deep understanding and improvement of novadays non-trivial business processes.
Usually, the traditional BPM approach implies obtaining data for the process modeling through interviews with participants. This inevitably puts their subjective views into the model. Because process mining works with actual data, it can solve this problem.

Process Mining techniques involve three main groups:
  • Process discovery
    Process discovery yields a process model based on event log.
  • Conformance checking
    Conformance checking links event log to the process model and compares both. The goal is to find similarities and differences between the observed behavior and the modeled behavior.
  • Model enhancement
    Model enhancement is the first step towards process enhancement. The goal is to help organizations maintain a high level of standardization, predictability and continual improvement of the processes.
Why use process mining?
Process Mining helps to find well-grounded solutions to real business problems that organizations often face in operational management.
  • Identifying process bottlenecks
    Process mining helps to find previously undetected bottlenecks and address their root causes to restore process efficiency.
  • Tracking process SLAs
    Process mining is useful in tracking of SLAs compliance, and can help to find and substantiate optimal values for service levels.
  • Customer satisfaction monitoring
    Process mining helps to monitor customer satisfaction in various forms (customer scores, extra purchases, recommendations, etc.) and identify dependencies on other process metrics.
  • Detecting process deviations
    Any organization knows how a business process should proceed. All process routes can be classified into normal and abnormal. Using digital footprint of the real process, process mining can monitor all stages and detect any deviations. This knowlege opens up wide opportunities to process improvement.
Where to begin?
Process Mining can be implemented on-premise or used as a service. On-premise model requires additional cost for building a process mining Center of Expertise (CoE) and IT infrastructure, including servers and specialized software. Process mining as a Service provides a quick start with minimal cost. In case of any doubt can be performed a pilot project.

As a first step towards implementing Process Mining in business, we suggest using Process Mining as a Service. How can we help? Process Mining combines several professional areas - applied mathematics, computer science, process approach to management, industry specifics of the company and its processes. To use our Process Mining as a Service, you just need to know the industry specifics. We will do the rest and present the results in the language of business.
How it works
In case of Process Mining as a Service we do almost all technical work in the project. As an outsourcer, we use our process mining CoE, including the work of business analysts, data engineers and data scientists, as well as IT Infrastructure. The customer participation is required at the start - together we define business requests, problems, hypotheses, agree on the initial data, and at the finish line - we discuss quantitative and qualitative results, develop a process improvement plan.
Define business requests, problems, hypotheses
[customer and outsourcer]
Agree on the initial data
[customer and outsourcer]
Create data model and perform ETL procedures
Select methods, create models, perform analysis
Prepare conclusions and recommendations
Discuss results and process improvement plan
[customer and outsourcer]
Implementation of Process mining requires specialized software. We use our own process mining software - Astelica Process Analyzer.
What is next?
We can also help with the next steps to implement process mining. We are ready to design the architecture and models for regular using, deploy and customize the necessary software solutions, and teach the independent use of process mining techniques.
Astelica Process Analyzer
process mining software
Astelica Process Analyzer is a specialized software for Process Mining, which is fully business-oriented and supports veritable data-driven management. It has a modular architecture and contains two main modules.
Module #1 performs extracting and pre-processing of raw data, which can be fully or partially unstructured and presented in various formats (csv, xlsx, SQL and NoSQL databases).

It is known that ETL for process mining is a non-trivial problem because ETL procedures must consider the specifics of each customer, features of his tasks and data. We take into account all of this one in module#1.

As a result of ETL procedures, we receive data in SQL format suitable for analysis by process mining methods. Note that we use a data model with an extended set of event and case attributes, not just "case_id - timestamp - activity".
Module #2 contains the main functionality of process mining. It allows analysts to flexibly form samples, obtain comprehensive descriptive statistics, evaluate the repeatability of the process, identify anomalies and relationships in the process, monitor compliance with SLAs and protocols, perform detailed case analysis, compare routes of the process executing, assess the staff workload, build predictive models.

Below are examples of analytical tools.
Multidimensional filtering of cases and forming the sample for analysis
Dashboard and descriptive statistics for routes, cases, events
Evaluation of the process repeatability and the feasibility of its automation
Assessment of frequency and duration for process execution routes
Study of the relationship between сlient grades and case execution time (it can help with SLA definition)
Assessment of staff workload through the dynamics of cases in progress
Benefits of Astelica Process Analyzer
  • End-to-end process representation
    Integration of data from different sources.
    Manual or automatic mode.
  • Solving real business problems
    SLA compliance. Customer satisfaction. The feasibility of automation. Workload and staff efficiency.
  • Process based analytics
    Deep specialization on processes. Methods of Process Mining. Data Science and Machine Learning in action.
  • Data-driven decision making
    Unbiased data from digital footprint of the process. Clear facts and figures for decision-makers.
Astelica Process Analyzer use cases
Astelica Process Analyzer software has already been used to analyze real business processes from various areas – housing and communal services, construction, outsourcing contact centers.
Business Process Management Consulting
traditional approach to business process modeling
Our company has considerable experience in traditional Business Process Management (BPM) consulting. Astelica Consulting specialists are engaged in description, modeling, optimization, unification of business processes in IDEF0, BPMN, EPC notations.

In our consulting projects, we work with people who are participants in the process, such as managers, process owners, process executors, obtain information about the process through interviews, and help the customer set up the process management system.

Currently, we are actively combining the capabilities of traditional BPM consulting approaches with Process Mining in projects. This provides a more objective and full representation of the business process.
Why use BPM consulting?
Consulting provides an outside perspective on your business processes and process management system.
  • Process management diagnostics
    To build development plans in process management, it is very important to understand the current maturity level of both individual business processes and process management system. External consultants have proven maturity models and can do so objectively.
  • Modeling of main end-to-end processes
    To model the main end-to-end processes, coordination of various functional units in the organization is necessary. Help of consultants can be useful in this.
  • Unification of processes in the holding
    Unification of processes in the holding requires alignment of different opinions and practice. Consultants can help to agree.
  • Transfer of best practices
    Consultants carry out projects in various industries. They can transfer best practices between industries and bring additional competitive advantages.
About Astelica Consulting
some facts about us
  • Business experience
    Before consulting, we worked in business for about 20 years in middle and senior positions. We know inside the banking, telecommunications, manufacturing and trading businesses.
  • Suitable base education
    Our consultants have basic education in physics and mathematics plus additional education in computer science, economics, finance, and management. This allows you to combine a systems approach and applied knowledge and skills in solving problems.
  • University teaching experience
    Some of our consultants are part-time university lecturers. They teach students in the specialty Business Informatics disciplines related to business process management, data science and process mining.
  • Smart applying of information technology in business
    We have been working with information technology for a long time and we know a lot about business digitalization. In our opinion, IT in business should be exactly as much as needed. Hype around any IT solution is not a reason to implement it.
  • Experience in data science and machine learning
    We have been applying data science and machine learning techiques in science and business for many years. These are solutions in the field of process mining, learning analytics, portfolio management of financial assets, HFT trading, etc.
Contact us by email