CORPORATE PROCESS ANALYSIS SYSTEM
Patent Information
- Application Number
- TR202612963
- Authority / Receiving Office
- TR · TR
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2026-07-31
- Publication Date
- 2026-09-21
Smart Images
Figure 00000018_0000
Abstract
Description
1 TARIFF CORPORATE PROCESS ANALYSIS SYSTEM Technical Area This invention automatically detects delays occurring in corporate business processes. Identifying and causally analyzing the factors that cause delays. determined using methods, learning from past process data With the help of artificial intelligence models, delay risks can be predicted in advance. 10 automated or user-supported solutions for estimating and determining delays It relates to a system that enables the generation of action recommendations. Previous Technique 15 developed for monitoring corporate processes in the known state of the art. Software solutions measure the completion times of processes, defined service reporting whether delays have occurred according to service level agreements (SLAs). or provides process performance indicators to the user. However, In current solutions, only information that a delay has occurred is obtained, and at which process step did the delay begin, and from which user or 20 it stems from the organizational structure, under what operational conditions the system has experienced this and there is a possibility that similar delays will occur in the future. This cannot be determined by them. Patent number US11663051B2, which is included in the prior art, 25 the document discusses optimizing workflows using machine learning. A method for this is explained. The steps involved in this solution process are... Modifying task planning by analyzing waiting times It provides. However, the user who caused the delay in the relevant patent. determining behaviors, making causal inferences, 30 in natural language any explanation regarding the generation of automated action suggestions 2 There is no technical solution. Brief Description of the Invention One aim of this invention is to reduce delays in corporate processes to just 5 Instead of determining it based on time duration, the user who caused the delay transaction type, organizational structure, time information, and process dependencies all together. a process that automatically identifies the root cause of the delay by analyzing it The goal is to develop a management system. Another purpose of this invention is to train machines on past process records. learning models to detect delay before it even occurs The goal is to develop a predictive analysis mechanism that enables the calculation of risk. Detailed Description of the Invention 15 The "Enterprise Process Analysis" carried out to achieve the purpose of this invention The "System" is shown in the attached figure; Figure 1. Schematic view of a system that is the subject of the invention. 20 The parts shown in the figure are individually numbered, and these numbers correspond to... The corresponding answers are given below. 1. System 25 2. Electronic device 3. Database 4. Server D. External server 30 3 Identifying delays in organizational processes, addressing these delays Identifying the causes, conducting risk analyses before delays occur. implementation and automated action suggestions regarding identified delays The system that makes it possible to create the subject of the invention (1); - 5 user requests for analysis of corporate processes entry, viewing process statuses, reports enabling the viewing and monitoring of the generated analysis results. at least one electronic device configured to provide a user interface device (2), - on it, corporate process records, start and end dates for process steps 10 their times, user transaction logs, organizational information, processes definitions, transaction types, service level agreements (SLAs), performance indicators (KPIs), process attribution information, delays records, causal analysis results, machine learning model parameters, risk scores, action recommendations, user feedback 15 notifications, system logs, explainable AI outputs to store other data and analysis results generated by the system at least one database structured to (3) and - via an external server (D) in the form of an external corporate information source obtaining data related to corporate processes, analyzing the obtained data 20 performing data preprocessing operations, processes in time. to correlate, identify delays, cause delays Identifying the contributing factors and predicting potential future delays. to develop action plans regarding the identified delays, to receive user feedback and 25% of the user feedback received The results created by learning are sent to the user via electronic device (2). It includes at least one server (4) configured to serve. The system in question consists of (1) electronic devices (2), smartphones, tablet computers or a device in the form of a portable computer. The electronic device in question (2) 30 any remote communication protocol included in the known state of the art 4 using to connect to server (4) and communicate data with server (4) It is structured to achieve this. In the preferred application of the invention. electronic device (2) server (4) using Internet as data bus data It is structured to carry out the shopping. Electronic device (2), 5. The user enters a request for process analysis regarding an organizational process. It is structured to provide a user interface that enables electronic communication. device (2), process analysis by user, delay detection, causes of delay, process inquiry, reporting, risk analysis, viewing of action recommendations It is structured to provide a user interface that enables electronic communication. Process analysis performed by device (2), server (4), resulting in 10 a system that allows outputs to be displayed to the user graphically and textually. It is configured to provide an interface. In the system that is the subject of the invention, the database (3) in (1) is in communication with the server (4) and is configured to be managed by the server (4). The invention is preferred 15 In the implemented application, the database (3), the process related to corporate processes their identities, process steps, process start and end time information, user their identities, user roles and organizational information, transaction types, requests, approvals, Purchase, contract, delivery, invoice and similar process records, event records, process transition information, process dependency data, service level agreement (SLA) 20 parameters, performance indicators (KPIs), work schedules, shifts and official holiday information, user workload data, process correlation results, normalized datasets obtained as a result of data preprocessing, missing data Completion results, delay detection results, delay classification records, results of statistical analysis regarding delay, causal analysis 25 cause-and-effect relationship records, process risk scores, and delay probability with the outputs. predictions, explainable AI analysis results, natural language explanation texts, action recommendations, automated action logs, and analysis outputs It is configured for storage. The server (4) in the system (1) which is the subject of the invention, any remote communication to communicate with the electronic device (2) using the protocol and the established protocol to exchange data with electronic devices (2) via communication is being configured. The server (4) in question enters new data into the database (3). registration, deletion of registered data in database (3) or data 5 Modifying the registered data in the database (3), within the database (3) Data processing involves operations such as updating and retrieving recorded data. It is configured to manage the base (3). Server (4), electronic device (2) to receive analysis requests for corporate processes transmitted through this channel and Depending on the request from the user, the Server (4) provides corporate information resources 10 process IDs, process steps, via an external server (D) in the form of, start and end timestamps, user IDs, organization information, transaction types, process logs, requests, approvals, purchases, deliveries, contract and billing records, service level agreement (SLA) parameters, performance indicators (KPIs), user workload information, calendar data, 15 to obtain shift information and official holiday records, and to enter the received data into the database (3) to record and process data and perform analysis operations on the data stored in the database (3) It is configured to be carried out via the server (4), corporate information There is a deficiency in the data received from an external server (D) in the form of resources. Data completion (Missing Value Imputation), outlier detection (Outlier 20 Detection), data cleaning, data normalization (Min-Max Normalization, Z-Score Normalization), data standardization (Standard Scaling), timestamp conversion (Timestamp Standardization), data Record linkage, data enrichment, and process. 25 to perform process correlation operations It is structured. The server (4) is a common data that can analyze process data. transforming it into a structure, ensuring data consistency, and in subsequent analysis stages. It is configured to prepare for use. Server (4), process Recreating the process flow by arranging the steps in chronological order, processes 30 Event correlation, process mining, event 6 Event log mining, object-oriented process mining (Object-Centric Process Mining - OCPM) and process discovery (Process Discovery) The server (4) is configured to use at least one of the methods. It is structured to calculate the completion times of the process steps. Server (4) monitors the entire process step, enterprise resource planning (ERP) system, business 5 process management (BPM) system, electronic tendering system, workflow management system through at least one external server (D) in the form of other corporate information resources compare with the received service level agreement parameters and delay It is structured to determine the processes that create the server (4), delay 10 to identify abnormal process behaviors during detection The server (4) is configured to exclude public holidays and weekends during latency analysis. endings, shift changes, working hours, user permissions, organization changes, task handover records, system maintenance times, and processes to perform contextual delay filtering by taking exceptions into account It is structured. The server (4) records the previous 15 events that occurred within the process chain. calculating the impact of delays on subsequent process steps, chain reaction. Identifying delays, analyzing delay propagation, and process dependencies. It is configured to generate the coefficients. The server (4) detects delay. user density, organizational structure, and process steps involved type, process complexity, process dependencies, number of concurrent tasks, process 20 volume, user performance indicators, department density, tasks distributions, time-based transaction densities, past process behaviors, processes By analyzing completion times and process transition information together, we can identify delays. It is configured to determine the variables that cause it. Server (4), direct and indirect relationships between variables causing delay 25 To identify, model cause-and-effect relationships, and create dependency structures. and a Bayesian network to perform root cause analysis, Bayesian belief network, structural causal model Causal Model (SCM), Directed Acyclic Graph (Directed Acyclic Graph - DAG), do-calculus, counterfactual analysis (Counterfactual 30 Causal impact analysis, Granger causality. 7 analysis (Granger Causality), decision tree, regression analysis (Regression Analysis), Random Forest Feature Importance (Importance), gradient boosting decision trees, SHAP Interaction Values and explainable artificial intelligence. (Explainable Artificial Intelligence - XAI) algorithms, at least one of 5 It is configured to use. The server (4) only has a delay. Identifying the previous process steps that triggered the delay, rather than the step in which it occurred. To identify critical bottlenecks in the process and create a delay propagation map. It is configured to create. The server (4) can create more than one in the same process. If there are reasons for excessive delay, these reasons are considered important. to rank them according to their degrees, to create an impact factor for each cause, and It is structured to calculate the contribution rates to the delay. The server (4) can detect future delays by using past process records. to train machine learning models in order to predict risks It is configured. The server (4) uses a random forest 15 for latency estimation. (Random Forest), Extreme Gradient Boosting (XGBoost), Gradient Boosting Machine (GBM), for long and short duration At least one of the algorithms in the form of a memory network (Long Short-Term Memory - LSTM) to calculate process delay probabilities using one is configured. The server (4) checks the accuracy of machine learning models 20 To improve, hyperparameter optimization, model selection, cross-referencing validation (Cross Validation), feature selection, feature Performing at least one of the dimensionality reduction operations It is structured in such a way. The server (4) is used in machine learning models. User-based processing intensity for generating the attributes to be used, 25 Number of simultaneous tasks per user, average processing time, average waiting time. duration, process cycle time, number of process steps, process complexity coefficient, organizational workload, department-based workload, hourly transaction volume distribution, daily transaction distribution, weekly transaction frequency, monthly transaction frequency, task completion rate, process dependency coefficients, user performance 30 to calculate indicators and past delay frequencies 8 The server (4) is configured to normalize the generated attributes. scaling, determining importance levels, and in machine learning models to create usable qualified feature sets is structured. The server (4) makes the generated prediction results explainable. Explainable Artificial Intelligence (Explainable Artificial Intelligence - 5) The server (4) is configured to use XAI algorithms. using explainable AI results that cause delays Listing the variables in order of importance shows the contribution each variable makes to the delay. to calculate the contribution and generate understandable explanations for the user. is structured. The server (4) has 10 natural language descriptions generated. to specify at which process step the delay occurred is being configured. Server (4) the main cause of delay, factors affecting delay variables, expected completion time, estimated delay time, and suggested The Server (4) analyzes to present corrective actions to the user in natural language. Converting the results into user-understandable natural language descriptions 15 for the purpose of natural language generation (NLG) and templates template-based text generation algorithms It is configured to use the server (4), using the analysis results. Creating a dynamic risk score for each process, classifying processes as low risk, medium risk, and classifying at high risk levels, exceeding the defined risk thresholds 20 to generate early warning in the event of a delay and inform the user before a delay occurs. It is configured to inform. The server (4) is configured to inform all pre-processing activities. Recording the processing, analysis and delay detection results into the database (3), Relating these results to past analyses, in subsequent analyses to use as reference data and to re-evaluate machine learning models 25 It is structured to be stored for use in training. Industrial Application of the Invention The system in question (1) is used for monitoring corporate processes and detecting delays. the analysis of the causes of delays with the help of artificial intelligence and 30 9 in information processing infrastructures for creating appropriate actions It can be implemented. System (1), electronic device (2), database (3) and server (4) using process logs, timestamps, user transaction data, analyzing performance indicators and other corporate process data; 5 in identifying delays, with causal analysis and machine learning algorithms to estimate the risks of delay and the results obtained electronic device (2) By presenting it to the user, it records it on the database (3). The subject of the invention is enterprise process management systems, enterprise resource planning. systems, workflow management systems and similar computing infrastructures It is applicable. 10 Within the framework of these fundamental concepts, the subject of the invention is the "Corporate Process Analysis System". It is possible to develop various regulations regarding (1)", and the invention is here The scope of protection of the invention is not limited to the examples given. As defined in the requirements. 15
Claims
REQUESTS 1. Identifying delays in organizational processes, Identifying the causes of delays, identifying risks before a delay occurs. performing analyses and automatically identifying delays 5 enabling the creation of action proposals; - the user enters process analysis requests, process viewing their status, viewing reports, and a user who enables monitoring of the generated analysis results at least one electronic device configured to provide the interface (2), 10 - contains corporate process records, including the starting and ending points of the process steps. expiration times, user transaction logs, organizational information, process definitions, process types, service level agreements (SLAs), performance indicators (KPIs), process attribution information, delay logs, causal analysis results, machine learning 15 model parameters, risk scores, action recommendations, user feedback, system logs, explainable artificial intelligence outputs and other data and analysis generated by the system at least one database structured to store the results (3) and - electronic device (2), database (3) and at least one external corporate information 20 to communicate with the system (5), process data from different data sources obtaining, data preprocessing operations on the obtained data to implement, to relate processes along a time axis, Identifying delays, identifying the causes of delays Identifying the factors, predicting potential future delays. 25 to develop action plans regarding the identified delays, learning from user feedback and the results generated configured to deliver to the user via electronic device (2) a system characterized by containing at least one server (4) (1). 11 2. Smartphone, tablet computer, desktop computer or portable Claim characterized by a computer-like electronic device (2) A system like the one in 1 (1).
3. Connect to the server (4) using any remote communication protocol 5 to establish and exchange data between the server (4) and the established connection. electronic device configured to enable the purchase to take place (2) A system like the one in Claim 1 or 2 characterized by (1).
4. Connecting to the server (4) via a data network such as the Internet 10 Claim 3, characterized by the electronic device (2) configured to do so. such a system (1).
5. Process analysis by the user, delay detection, causes of delay, process inquiry, reporting, risk analysis, action recommendations 15 electronic system configured to provide an interface that allows viewing. any of the above claims characterized by the device (2) such a system (1).
6. Process analysis performed by Server (4), resulting in 20 enables the outputs to be displayed to the user graphically and textually. an electronic device (2) characterized by being configured to provide an interface a system like any of the above-mentioned requests (1).
7. Communicating with Server (4) and being managed by Server (4) 25 from the above requests characterized by the structured database (3) a system like any other (1).
8. Process IDs relating to corporate processes via Server (4), process steps, process start and end time information, user IDs, 30 user role and organization information, transaction types, request, approval, purchase 12 process records such as receipt, contract, delivery, invoice and similar records, event records, process transition information, process dependency data, service level agreement (SLA) parameters, performance indicators, work schedules, shift and public holiday information, user workload data, process The correlation results are normalized from the data preprocessing results. processed datasets, missing data completion results, delay detection results, delay classification records, statistics on delays analysis results, causal analysis outputs, and cause-and-effect relationship records, training data used in training machine learning models clusters, model parameters, attribute data, process risk scores, 10 delay probability estimates, natural language description texts, action storage of recommendations, automated action logs, and analysis outputs. characterized by the database structured to provide (3) a system like any of the above requests (1).
9. Using any remote communication protocol, with the electronic device (2) to communicate and to obtain data through electronic devices (2) via this communication characterized by the server (4) configured to carry out the transaction. a system like any of the above-mentioned requests (1).
10. Server configured to receive process analysis requests transmitted through it. (4) like any of the above-mentioned claims characterized by system (1).
11. Depending on the user's request, 25 entered via the electronic device. depending on the process analysis request, in the form of corporate information resources Process IDs, process steps, start and end points are retrieved from the external server (D). expiration timestamps, user IDs, organizational information, transaction types, process logs, request, approval, purchase, delivery, contract and billing records, service level agreement (SLA) 30 parameters, performance indicators (KPIs), user workload 13 information, calendar data, shift information, and official holiday records to receive, record the received data into the database (3) and data processing and analysis to perform operations on data stored in the database (3) The above is characterized by the server (4) configured to do so. a system like any of the requests (1). 5 12. Obtained from an external server (D) in the form of corporate information resources. data completion, outlier detection, and data cleaning. data normalization, data standardization, timestamp conversion, data matching, data enrichment, and process correlation operations 10 characterized by the server (4) configured to perform a system like any of the above requests (1).
13. Transforming process data into an analyzable common data structure, data 15 to ensure consistency and use in subsequent analysis stages The above is characterized by the server (4) configured to prepare for the above. a system like any of the requests (1).
14. On the data received from the external server (D) in the form of corporate information Missing data completion, outlier detection, data cleaning, data 20 normalization, data standardization, timestamp conversion, data matching, data enrichment, and process correlation operations characterized by the server (4) configured to perform a system like any of the above requests (1).
15. Calculating the completion times of the process steps, the entire process step, Enterprise resource planning (ERP) system, business process management (BPM) system, electronic tendering system, workflow management system and other corporate service received via at least one external server (D) in the form of information resources compare with the level agreement parameters and the 30 that creates a delay 14 characterized by the server (4) configured to determine the processes a system like any of the above requests (1).
16. Identifying abnormal process behaviors during delay detection, delay During the analysis, official holidays, weekends, shift changes, 5 working hours, user permissions, organizational changes, tasks taking into account transfer records, system maintenance times, and process exceptions configured to perform contextual delay filtering by taking any of the above requests characterized by the server (4) such a system (1). 10 17. Previous delays occurring within the process chain affect subsequent processes. calculating the impact on steps, identifying chain delays, delay to analyze its spread and to create process dependency coefficients 15 of the above requests characterized by the configured server (4) a system like any other (1).
18. User traffic related to process steps where delays were detected, organizational structure, type of operation, process complexity, process dependencies, peers number of scheduled tasks, transaction volume, user performance indicators, 20 departmental workload, task assignments, time-based processing loads, past process behaviors, process completion times, and process transitions By analyzing this information together, we can identify the variables causing the delay. The above is characterized by the server (4) configured to determine the above. a system like any of the requests (1). 25 19. Direct and indirect relationships between variables causing delays. to identify, model cause-and-effect relationships, and determine dependency structures. To create and perform root cause analysis, Bayesian networks are used. belief network structural causal model directed non-cyclic graph, do-30 computational versus factual analysis, causal effect analysis, Granger causality decision tree analysis, regression analysis, random forest feature significance Gradient-increasing decision trees, SHAP interaction, and explainable artificial intelligence. with a server configured to use at least one of the algorithms (4) a system like any of the above characterized claims (1). 5 20. Not only the step in which the delay occurred, but also the preceding factors that triggered the delay. Identifying process steps, removing critical bottlenecks within the process. and the server (4) configured to generate the delay propagation map A system like any of the above characterized claims 10 (1).
21. In cases where there are multiple reasons for delay within the same process, the statement The topic is to rank the causes in order of importance, and to determine the impact of each cause. to create the coefficient and calculate the contribution rates on the delay 15 The above is characterized by the server (4) configured to do so. a system like any of the requests (1).
22. Using past process records to identify future delays training machine learning models to predict risks 20 The above is characterized by the server (4) configured to do so. a system like any of the requests (1).
23. Random forest and extreme gradient for calculating process delay probabilities. boosting, gradient boosting machine, long-term and short-term memory network, in the form of 25 server (4) configured to use at least one of the algorithms a system like any of the above characterized claims (1).
24. Hyper-30 to improve the accuracy of machine learning models parameter optimization, model selection, cross-validation, feature selection, 16 to perform at least one of the feature reduction operations from the above requests characterized by the configured server (4) a system like any other (1).
25. Creating features to be used in machine learning models. 5 In order to achieve this, user-based transaction intensity, simultaneous tasks per user number, average processing time, average waiting time, process cycle time, number of process steps, process complexity coefficient, organizational density, Department-based transaction volume, hourly transaction distribution, daily transaction volume distribution, weekly trading frequency, monthly trading frequency, task completion 10 ratio, process dependency coefficients, user performance indicators, and with server (4) configured to calculate past delay frequencies a system like any of the above characterized claims (1).
26. Normalizing, scaling, and assigning importance to the generated attributes. to determine their levels and use them in machine learning models with the server (4) configured to create qualified feature sets a system like any of the above characterized claims (1). 20 27. In order to make the generated forecast results explainable. Server configured to use explainable artificial intelligence algorithms (4) like any of the above-mentioned claims characterized by system (1). 25 28. Using explainable AI results to identify the cause of the delay. Listing the variables in order of importance, assigning a delay to each variable. to calculate the contribution made and provide understandable explanations to the user. 30 characterized by the server (4) configured to create a system like any of the above requests (1). 17 29. At which process step does the delay occur in the generated natural language descriptions? to state that it occurred, the underlying cause of the delay, factors affecting the delay variables, expected completion time, estimated delay time, and 5. To present the suggested corrective actions to the user in natural language. from the above requests characterized by the configured server (4) a system like any other (1).
30. Translate the analysis results into user-understandable natural language descriptions. Natural language generation and template-based text generation for the purpose of transformation 10 characterized by the server (4) configured to use its algorithms. a system like any of the above-mentioned requests (1).
31. To create a dynamic risk score for each process using the analysis results. classifying processes into low risk, medium risk, and high risk levels, 15 to generate early warning when defined risk thresholds are exceeded and to inform the user before any delay occurs from the above requests characterized by the configured server (4) a system like any other (1).
32. Data containing all preprocessing, analysis, and delay detection results performed. recording the results into the base (3) with past analyses to correlate, and to use as reference data in subsequent analyses. for use in retraining machine learning models 25 of the above requests characterized by the configured server (4) a system like any other (1).