Genetic Process Mining Research Articles

Timed Genetic Process Mining for Robust Tracking of Processes under Incomplete Event Log Conditions

In process mining, an event log is a structured collection of recorded events that describes the execution of processes within an organization. The completeness of event logs is crucial for ensuring accurate and reliable process models. Incomplete event logs, which can result from system errors, manual data entry mistakes, or irregular operational patterns, undermine the integrity of these models. Addressing this issue is essential for constructing accurate models. This research aims to enhance process model performance and robustness by transforming incomplete event logs into complete ones using a process discovery algorithm. Genetic process mining, a type of process discovery algorithm, is chosen for its ability to evaluate multiple candidate solutions concurrently, effectively recovering missing events and improving log completeness. However, the original form of the genetic process mining algorithm is not optimized for handling incomplete logs, which can result in incorrect models being discovered. To address this limitation, this research proposes a modified approach that incorporates timing information to better manage incomplete logs. By leveraging timing data, the algorithm can infer missing events, leading to process tracking and reconstruction which is more accurate. Experimental results validate the effectiveness of the modified algorithm, showing higher fitness and precision scores, improved process model comparisons, and a good level of coverage without errors. Additionally, several advanced metrics for conformance checking are presented to further validate the process models and event logs discovered by both algorithms.

Process Mining on New Student Admission Process in Telkom University using Genetic Miner

The selection process for new students at Telkom University, also known as SMB Telkom University has been running for years and already has its process flow. However, the existing process flow can be further improved to better reflect the actual field processes and become more accurate. Process mining can enhance this process flow by creating a new process flow based on event logs or previously executed processes. One of the algorithms in process mining is genetic process mining, where process mining is performed multiple times over several generations and genetic algorithms such as crossover and mutation are applied to generate a more accurate process model compared to other process mining algorithms such as heuristic and inductive mining. After conducting experiments, the best process model that was produced was at the 100th generation which has a fitness point of 0.755910819 and precision point of 0.742857143, after examining the parameters and the resulting Petri net or process flow that was produced it was concluded that the process model obtained from the application of Genetic Process Mining to SMB Telkom University is not very good because the resulting Petri net has several duplicate activities and appears to be non-linear. This could be due to several factors i.e., incompatible, or inaccurate data.

A multi-objective medical process mining model using event log and causal matrix

Medical analytics in genetic process mining have developed models with multiple and often conflicting criteria. Some studies have reduced the complexities inherent in multiple and conflicting criteria by eliminating some criteria and solving the problem with a single or linear objective function. This study proposes a multi-objective model for obtaining Pareto without removing any conflicting criteria. Two multi-objective Pareto-based models of Non-dominated Ranked Genetic Algorithm (NRGA) and Non-dominated Sorting Genetic Algorithm (NSGAII) are developed. A comprehensive examination and statistical analysis are performed to analyze the performance of the two algorithms with five metrics of Mean Ideal Distance (MID), Cover Surface (CS), diversity, spacing, and time. In addition, sensitivity analysis examines the problem size and the coefficients of the different criteria on the objective functions. Finally, for the performance evaluation, a Data Envelopment Analysis (DEA) is developed based on some instances and benchmarks to assess the efficiency of the algorithms. We show that NSGAII outperforms the NRGA according to the performance metrics and statistical hypothesis testing.

Combining metaheuristics and process mining: Improving cobot placement in a combined cobot assignment and job shop scheduling problem

Human workers can share a workspace with modern collaborative robots (cobots). The main differences to traditional robots are, that workers do not need a safety distance when interacting with cobots, as they move slower than typical industrial robots. Cobots also have fast setup times compared to traditional resources. The hybridization is based on a previous work of the authors, where a job shop scheduling problem that is extended with a robot to workstation assignment with a hybrid genetic algorithm is solved. In this paper, the potential of hybridizing an optimization algorithm with process mining techniques to improve the solution quality by gaining information on the solution structure is analyzed. This additional information should help to guide the search process. Process mining techniques are presented to analyze the solutions and learn from them. The idea of this work is to understand the solutions generated by the genetic algorithms as process executions, e.g., the production of a part as a process instance executed across the selected work stations. Then, by generating process event log data out of selected solutions, state-of-the-art process mining techniques can be used as visualization and scanning tools for the underlying processes. This way, for example, bottleneck workstations in the production process can be highlighted. Based on created scenarios, this paper demonstrates how genetic algorithms and process mining techniques can be combined. In the future, it is planned that information that is mined from generated logs of an evaluation framework is used to improve the performance of hybrid genetic algorithms by using this information in a feedback loop. Generated insight in cobot placement can also be used for prescriptive analytics in real-world manufacturing companies that want to utilize cobots. The focus of this paper lies in the discussion of the usage of potential information extracted from process mining.

A slippery genetic algorithm-based process mining system for achieving better quality assurance in the garment industry

Due to the error-prone nature of garment manufacturing operations, it is challenging to guarantee the quality of garments. Previous research has been done to apply fuzzy association rule mining to determine process settings for improving the garment quality. The relationship between process parameters and the finished quality is represented in terms of rules. This paper enhances the application by encoding the rules into variable-length chromosomes for optimization with the use of a novel genetic algorithm (GA), namely the slippery genetic algorithm (sGA). Inspired by the biological slippage phenomenon in DNA replication, sGA allows changes to the chromosome lengths by insertion and deletion. During rule optimization, different parameters can be inserted to or removed from a rule, increasing the diversity of the solutions. In this paper, a slippery genetic algorithm-based process mining system (sGAPMS) is developed to optimize fuzzy rules with the aim of facilitating a comprehensive quality assurance scheme in the garment industry. The significance of this paper includes the development of a novel variable-length GA mechanism and the hybridization of fuzzy association rule mining and variable-length GAs. Though the capability of conventional GA in rule optimization has been proven, the diversity in the population is inherently limited by the fixed chromosome length. Motivated by this phenomenon, the sGA suggested in this paper allows various parameters to be considered in a rule, improving the diversity of the solutions. A case study is conducted in a garment manufacturing company to evaluate the sGAPMS. The results illustrate that better quality assurance can be achieved after rule optimization.

A hybrid approach to extract business process models with high fitness and precision

Process mining (PM) aims at extracting a process model from an event log to represent the process behavior recorded in that event log. An extracted process model with high fitness and precision means it can reflect most of the process behavior recorded in the event log (fitness) and will not generate extra behavior not recorded in the event log (precision). Most of the existing PM methods, such as the genetic process mining (GPM), focused only on the achievement of high fitness, but ignore the pursuit of high precision. This research presents a hybrid PM approach that integrates the GPM, particle swarm optimization, and differential evolution to extract process models with high fitness and precision (FP values) from event logs. The results show that the proposed approach achieves improvement in extracting process models from event logs with higher FP values.

An Integrated mining approach to discover business process models with parallel structures: towards fitness improvement

Process mining (PM) is a technique to extract a process model from an event log to represent the process behaviour recorded in that event log. A mined process model with high fitness means that it can reflect most of the process behaviour recorded in the event log. Previous studies have shown that the mined model with high fitness can be used in process improvement, such as fraud detection, continuous process improvement and benchmarking. Genetic process mining (GPM) is a famous PM approach, which can simultaneously identify several process structures from event logs. However, GPM cannot effectively discover parallel structures from event logs. This study proposes a PM approach based on integration of GPM, particle swarm optimisation and differential evolution to find process models with high fitness for event logs involving multiple parallel structures. The results show that the proposed approach does indeed lead to improvement in gaining process models with high fitness for event logs involving multiple parallel structures.

DETECTING ABNORMAL BEHAVIOR IN SOCIAL NETWORK WEBSITES BY USING A PROCESS MINING TECHNIQUE

Detecting abnormal user activity in social network websites could prevent from cyber-crime occurrence. The previous research focused on data mining while this research is based on user behavior process. In this study, the first step is defining a normal user behavioral pattern and the second step is detecting abnormal behavior. These two steps are applied on a case study that includes real and syntactic data sets to obtain more tangible results. The chosen technique used to define the pattern is process mining, which is an affordable, complete and noise-free event log. The proposed model discovers a normal behavior by genetic process mining technique and abnormal activities are detected by the fitness function, which is based on Petri Net rules. Although applying genetic mining is time consuming process, it can overcome the risks of noisy data and produces a comprehensive normal model in Petri net representation form.

The Behavior Trustworthiness Analysis Methods of Composite Web Services Based on Process Mining

At present, the behavior trustworthiness of the composite Web service was very important. The soundness of the composite Web service model was the core of the behavior trustworthiness of the service composition, which includes two steps, one was to determine whether or not the composite model satisfied the behavior relati vity, in order to analyze the function predictability; the other was to compare the behavior conformance for analyzing the influe nce of external factors . But the proposed judging algorithm of behavior relativity was feasible only in the case that T-invariant of Petr i net model exists, and the behavior conformance was also difficult to handle. In the paper, firstly, in order to analyze the behavior r elativity between the Web services, an algorithm for determining the behavior weak soundness of the composite service was presented based on the theory of the service tree. Then, we present an effective analysis method of behavior conformance based on process mining. Finally, we develop plug-ins and test our methods in large benchmarks, and compare our methods with genetic process mining methods. The theoretical analysis and experimental results indicate that this method can realize the predictable behavior of software we ll, and satisfy the behavior trustworthiness requirements of software.

The trustworthiness analyzing of interacting business process based on the induction information

Under the open environments, it is very difficult to guarantee the trustworthiness of interacting business process using traditional software engineering methods, at the same time, for dealing with the influence of external factors, some proposed business process mining methods are only effective 1-bounded business process, and some behavior dependent relationships are ignore. A behavior trustworthiness analysis method of business process based on induction information is presented in the paper. Firstly, aimed to the internal factors, we analyze the consistent behavior relativity to guarantee the predictable function. Then, for the external factors, in order to analyze the behavior change of business process, we propose a process mining methods based on induction information. Finally, experiment simulation is given out, and compares our method with genetic process mining methods. Theoretical analysis and experimental results indicate that our method is better than the genetic process mining method.

Time-interval process model discovery and validation—a genetic process mining approach

A process management technique, called process mining, received much attention recently. Process mining can extract organizational or social structures from event logs recorded in an information system. However, when constructing process models, most process mining searches consider only the topology information among events, but do not include the time information. To overcome the drawbacks, a time-interval genetic process mining framework is proposed. First, time-intervals between events are derived for all event sequences. A discretization procedure is then developed to transform time-interval data from continues type to categorical type. Second, the genetic process mining method which is based on global search strategy is applied to generate time-interval process models. Finally, a precision measure is defined to evaluate the quality of the generated models. With the measure, managers can select the best process model among a set of candidate models without human involvement.

Quantifying process equivalence based on observed behavior

In various application domains there is a desire to compare process models, e.g., to relate an organization-specific process model to a reference model, to find a web service matching some desired service description, or to compare some normative process model with a process model discovered using process mining techniques. Although many researchers have worked on different notions of equivalence (e.g., trace equivalence, bisimulation, branching bisimulation, etc.), most of the existing notions are not very useful in this context. First of all, most equivalence notions result in a binary answer (i.e., two processes are equivalent or not). This is not very helpful because, in real-life applications, one needs to differentiate between slightly different models and completely different models. Second, not all parts of a process model are equally important. There may be parts of the process model that are rarely activated (i.e., “process veins”) while other parts are executed for most process instances (i.e., the “process arteries”). Clearly, differences in some veins of a process are less important than differences in the main arteries of a process. To address the problem, this paper proposes a completely new way of comparing process models. Rather than directly comparing two models, the process models are compared with respect to some typical behavior. This way, we are able to avoid the two problems just mentioned. The approach has been implemented and has been used in the context of genetic process mining. Although the results are presented in the context of Petri nets, the approach can be applied to any process modeling language with executable semantics.

Genetic process mining: an experimental evaluation

One of the aims of process mining is to retrieve a process model from an event log. The discovered models can be used as objective starting points during the deployment of process-aware information systems (Dumas et al., eds., Process-Aware Information Systems: Bridging People and Software Through Process Technology. Wiley, New York, 2005) and/or as a feedback mechanism to check prescribed models against enacted ones. However, current techniques have problems when mining processes that contain non-trivial constructs and/or when dealing with the presence of noise in the logs. Most of the problems happen because many current techniques are based on local information in the event log. To overcome these problems, we try to use genetic algorithms to mine process models. The main motivation is to benefit from the global search performed by this kind of algorithms. The non-trivial constructs are tackled by choosing an internal representation that supports them. The problem of noise is naturally tackled by the genetic algorithm because, per definition, these algorithms are robust to noise. The main challenge in a genetic approach is the definition of a good fitness measure because it guides the global search performed by the genetic algorithm. This paper explains how the genetic algorithm works. Experiments with synthetic and real-life logs show that the fitness measure indeed leads to the mining of process models that are complete (can reproduce all the behavior in the log) and precise (do not allow for extra behavior that cannot be derived from the event log). The genetic algorithm is implemented as a plug-in in the ProM framework.