BPM trends

Business Process Management system (BPMS) represents a tool for implementing the concept of process management of an organization, considering business processes as special resources, which continuously adapt to constant changes; it relies on the principles of clarity and visibility of business processes, which are modeled using formal notations, simulation, monitoring and analysis of business processes, and the dynamic restructuring of business process models. Using the BPMS, data is collected, processes are modeled and their execution is monitored, as well as their logic is adjusted and configured depending on the needs of the company and the state of the external environment.


The use of BPM is associated with the tasks of enhancing the productivity, efficiency and competitiveness of organizational and production systems in all industries by increasing the efficiency of process activities, which leads to economic effects.


The development of BPMSs is associated with the concepts for describing business processes, the existing notations, and the standards, which allow to apply these systems in production systems.


One of the most common uses is BPMaaS (the Business Process Management as a Service), which allows you to automate both internal and external processes. About 60% of BPMSs is deployed on local infrastructure due to security issues; currently, the decisions on the organization of works and the use of internal resources become the major competitive advantage, since the distinction between the products manufactured by competitors is becoming smaller and smaller).


It is claimed, that BPMSs provide the basis for building an ecosystem for businesses. BPMSs play a significant role in implementing digital transformation strategies for businesses, helping them to identify bottlenecks and improve productivity.


Another trend in the BPMS development is intuitive process modeling, which is also ensured by the spread of low-code and no-code development offerings.

EdPM Modeler

No-code system is for the visual creation of structural-functional models and the automatic generation of R language simulation models. The program can be used to describe information, production, organizational, and other processes. For graphical representation, the EdPM/EPM notation is used, which allowed us to implement:


  • structural-functional modeling using graphical methods;
  • the study of the efficiency of structural-functional models using simulation methods, that allow (e.g. unlike Petri nets) to process queries in groups, which is important for the study of the efficiency of using such methods as volumetric calendar planning and AI methods in process activities, since the operating time of these methods depends on the number of parameters and changes nonlinearly;
  • the study of multiprocess systems;
  • the results were obtained, that allow you to find efficient topologies of structural-functional models.

Installation


To install and run the application, you need to download the archive (program file, icon, e[amples) – EdPM.zip (~2 MB). Compiled Java program containing all the necessary modules and components. To run the program you will need Java JDK of the latest version, which needs to be downloaded and installed.


To study models, using simulation modeling tools, you need to download and install the interpreter the R language, and if you need to refine and debug the resulting models - the R language development tool (e.g. R-Studio).


The program is distributed under license GPLv3. Source codes can be provided upon request.


To run the program under Windows, you need to create a folder for the program and execute the command

java -cp EdPM.jar EPM.mdi


To avoid the repetitive performance of the operation described above, you can create an EdPM.bat file with the following content:

---------------- file EdPM.bat start -----------------

@echo off
cd ./Programm Files/EdPM
java -cp EdPM.jar EPM.mdi

---------------- fail EdPM.bat end -----------------

Now you can place a link to this file in a convenient place and call the program by clicking on the corresponding icon.
To run the program on MacOS, you need to create a folder for the program and call it using the command

java -cp EdPM.jar EPM.mdi


To run the program on MacOS, you need to create a folder for the program and call it using the command

java -cp EdPM.jar EPM.mdi


To avoid the repetitive performance of the operation described above, you can create a .sh file with the contents above or create a Mac executable file, so you need to:
  1. Download and install the program Platypus.
  2. Create file EdPM.s

    3. ---------------- file EdPM.sh start -----------------

    #!/bin/bash

    cd /Applications/EdPM
    java -cp EdPM.jar EPM.mdi

    exit

    ---------------- file EdPM.sh end -----------------

  3. Download file with icons icon.icns (you can also create an icon for an application)
  4. Create an executable file using Platypus.


Usage demonstration


EdPM / EPM notation


The description of the main elements


Specification Feature Description Legend


The activation block (S)

 generates a prime natural number using a Poisson stream with a given probability or a periodically repeating value Action block activation, the generation of the number of requests/applications/calls S=prob()
S=periodic()


The action block (V)

 implements an operation that will result in data, or a delay associated with the computational complexity, and then the result will be queries that will appear at the output right after. Request processing R=V(,S,NV,O())
R=F(S,NV) – for the case of a self-implementing function


Result block (R)

 The storage and transmission of the result to the next block at the next clock cycle via the existing connection. Storing results of the action block (V) —


The Environment/Context Block (NV)

 The block stores data required for the operation of the Action block (V). Storing data necessary for the model to operate. NV=R
NV$=


The block responsible for performance (O)

 helps affect the speed of the Action block (V). It is required to take into account the influence of different types of equipment on the algorithm. — —


The condition block (V_if)

 is used to wait for actions to be performed (synchronization and control of the sequence of processes) The implementation of loops and conditional transitions do then else
Connection Data transfer Data transfer = – assigning a value to the block that the arrow enters
+ – data merging when multiple arrows occur
∩ – data extraction when combining the results of several passes in one block when organizing cycles

Publications


  1. Mylnikov, L.A., Saltykova, A.D. и Avramovic, Z. (2024) «Event-Driven Process Methodology Notation for Information Processing Research», Automatic Documentation and Mathematical Linguistics, 58(4), сс. 243–254. Available at: https://doi.org/10.3103/S0005105524700183.
  2. Mylnikov, L.A. и Suntsov, V.P. (2023) «Statistical Analysis of Simulation Data for the Evaluation of Information Processes’ Performance», Automatic Documentation and Mathematical Linguistics, 57(4), сс. 237–247. Available at: https://doi.org/10.3103/S0005105523040088.
  3. Mylnikov, L.A. (2022) «Structural and Functional Modeling of Processes with a Dedicated Control Subject», Automatic Documentation and Mathematical Linguistics, 56(1), сс. 42–54. Доступно на: https://doi.org/10.3103/S0005105522010083.

Structural-functional modeling

To understand how structural-functional modeling helps to describe and improve the efficiency of process activities, watch Tom Wujec's talk at the TedEx conference in June 2013.



For structural-functional modeling, many notations and software systems have been developed to enable it to be carried out (e.g. IDEF0, ARIS (BPMN и eEPC), TOGAF/ArchiMate, SysML). However, in order for the efficiency of the models to be studied by formal methods, there are only a few approaches, that have applicability limitations associated with either the need to post applications/requests one by one (Petri nets) or the need to already have an implemented system for collecting statistics (Process Mining), or the need to develop models in notations and descriptions that differ from the notations for the operation of the logic of individual elements (simulation modeling).

The approach, called ProcessMining, was proposed by a professor at RWTH Aachen University. Wil van der Aalst. The basic principles are described in his book Process Mining. Information about the direction, links to software and examples can be found on the website http://processmining.org.


The lecture by Prof. Wil van der Aalst - "How process mining improves the things you do not see".


This approach on computers became known after the works by Jay Wright Forrester who laid down the principles of modeling systems based on simulating resource flows and requests.


The principles he laid down are used in a large number of modern simulation systems: SciLab, Simulink, LabVIEW, OpenModelica.


The lecture by Prof. Jay W. Forrester - "Applications of System Dynamics".


The approach known as Petri nets was proposed byCarl Adam Petri. To get an idea of the proposed approach, you can read his article Petri net.


The most famous systems that allow you to explore the work of Petri nets: PIPE2, WooPeD, GreatSPN, YAWL.


Wolfgang Reisig - "Carl Adam Petri und seine Netze" (German).


Feedback

We will be glad to receive any suggestions for the development of the system's functionality, error messages, assistance in improving the system, the opportunity to test it on real cases, descriptions of examples of successful use, and other types of cooperation. To contact us, please use the feedback form:



Event-driven Process Methodology


           Copyright © 2024 Process-Simulation. All rights reserved.