Modified As-is Business Process Model and Notation
This report is a documentation of the modifications made to the as-is and to-be processes of the data mining process. The as-is and to-be BPMN are drawn using ADONIS, an online BPMN software. The report further expounds on what tasks and processes have been modified, why they have been modified and how the modifications will make the data mining process better than it was before the modifications that were done to the as-is and to-be processes.
With reference to (Börger 2012) and (Geamba?u 2012) this is the as-is BPMN for the data mining,
The existing as-is BPMN did not have the provision for checking whether the data provided was from a trusted source or whether it is valid to the results anticipated for the data mining process. In the modified as-is process there is the introduction of the validity checks to determine whether the data provided by the data provider is valid and is of required nature. If the validity checker determines that the data is not suitable or is not from a trusted source then the system takes the data back to the data provider for resubmission of trustworthy and valid data for storage by the data collector. If the data provided is valid then the data is forwarded to the data collector for storage.
In the business understanding phase there is the introduction of identification of the problem, before the business problem definition. This improvement will help reduce the data mining time as the problem has been identified and the next thing is to find the best way to solve the problem with the data miner. This, in turn saves on the time taken to conduct the data mining and the cost of conducting data mining process.
The provisions of data from the data provider should be checked for consistency, accuracy and the origin of the data to establish whether the data provide is of trusted origin.
For a business problem in the business understanding to be defined, the problem must first exist and be identified. The new Business Process Model Notation as-is diagram, identifies the business problem to be exploited, expounded and defined before the problem can be further looked into.
The checking of the validity of data will ensure that the data provided to the data collector for storage is accurate and correct. This means that the resultant results after the mining process will be accurate and reliable. Hence enabling the whole process to be reliable and accurate.
Modified To-be Process
In the business understanding phase there is the introduction of identification of the problem before the business problem definition. This improvement will help reduce the data mining time as the problem has been identified and the next thing is to find the best way to solve the problem with the data minor, which in turn saves on the time taken to conduct the data mining and the cost of conducting data mining process.
In the implementation of the data collection, there is the enforcement of checking whether the data collected is valid or the expected data. If the data is valid then it proceeds to the business understanding process and if the data is not valid it is send back for resubmission by the data provider (Chinosi and Trombetta 2012).
Just like in the as-is BPMN there is also the introduction of the identification of the business problem before expounding on it.
The data provided is checked to ensure that the data is accurate, consistent and original. To ensure that the data is not sourced from a null or source that will only cause problems during mining.
Before a problem is expounded on, the problem should first be identified. A non-existent problem cannot be defined before being identified. An identified problem is easier to identify as compared to an ambiguous problem or stating the definitions of many problems not closing to specifics.
Checking the validity of data ensures that the end results of data mining will be accurate and reliable. This enables achievement of the main aim of data mining which is the acquisition of the required accurate and reliable results.
Identifying the business problem at the data collection stage makes work easier for the data miner in the mining process where the problems won’t need to be identified anymore but defined thus reducing the total time taken to complete the data mining process
Identification and definition of the problem are removed from the BPMN. The identification and definition were placed as a responsibility of the data collector and cannot be repeated by the data miner.
The data miner is supposed to undertake the design of the models for data mining. The system then implements the model on the first sample data (Han, Pei & Kamber 2011) then validates on the second sub sample data. After validation the system then deploys the model for data analysis and produces a report which the data minor receives. The data minor upon receiving the report, he or she sends the report to the decision maker.
Modified To-be Business Process Model and Notation Diagram for the Decision making
The business problem identification and definition is done by the data collector therefore there is no need of repetition of the same task by the data miner. The repetition of the same by the data miner would be time consuming and slow the process of data mining, which could then take a lot of time to complete.
The three modeling steps in data mining have been implemented to address the accuracy of the data and improve on the time taken to complete data mining. This is made possible by reusing the available models already created.
The delegation of tasks and responsibilities helps reduce the time taken to complete the data mining process as some of the tasks won’t be recurring. For instance the problem identification and definition which was catered by the data collector need not be repeated by the data miner. The data miner therefore, observes data and design mining models. This enhances the time taken to complete the whole process of data mining.
The testing, validation and deployment of the data mining model ensures that the data is subjected to the correct data mining model. This ensures accuracy of the results gained after the data mining process. The process will also be fast and saves on time during the mining process as the model established will be reused on the rest sub-sample data. The implementation of the modeling steps by the system makes the process of data mining more accurate, efficient, effective and timely.
The decision in this case is made by the decision maker. In the prevoius system both the analysis and the decision was made by the system. In this case the system does the analysis and the decision maker does the decision making.
This is because the system is only supposed to do the analysis of the report provided by the data miner and then propose to the decision maker of the necessary decisions to be made. The decision maker taking into account the proposition then makes the decision.
The decisions made by the decision maker will be accurate, informed, consistent and timely. After receiving the analysis of the data provided by the data miner and the decision propositions from the system the decision maker is now able to provide accurate, informed, consistent and timely decisions.
- During data mining process, the system provides accurate, efficient, effective and timely data after the implementation of the data models on the data.
- During the implementation of the data models, the system is faster than the data mining person.
- The data miner could have received data that is both not original and inaccurate. That is why the role of the decision maker is to check whether the data provided by the miner is reliable, of required quality and original. This is done by tracing the original source of the data modified and reported to the decision maker.
- The problem identified and defined by the data collector is accurate and in line with the requirements of the data miner. That is why the data miner doesn’t have to look for, identify and define a business problem to be solved in the business understanding process but, focusses on observing the data, designing and defining mining models to accomplish the mining task.
- The data collected by the data collected is not always correct and has to be checked before it is stored and taken to the next process.
- The data miner could violate the agreement or contract of no disclosure between the decision maker and the miner. That is why the original source and the report has to be checked for leakage before the information is analyzed and used for decision making.
- There is no need for the data provider as the data collector can perform the activities done by the data provider. The data provider is only there as a formality, otherwise data is collected from the internet, a job the data collector can perform comfortably.
- The data miner could have received data that is both not original and inaccurate. That is why the role of the decision maker is to check whether the data provided by the miner is reliable, of required quality and original. This is done by tracing the original source of the data modified and reported to the decision maker.
This is with reference to the cost analysis xls document
The modification included the following personell; data collector, the data miner and system, the decision maker and the system. Before, the process included the data provider, collector, miner and the decision maker.
The data collector is assigned so much responsibility to ensure the data is accurate and to ease the work of the data miner . The data collector is also responsible for identifying the business problem to be defined. By doing so the data miner is left with defining the business problem and designing and implementing the various models to be used in data mining process.
The removal of the data provider helps to reduce the cost, as the data collector takes up the responsibilities of the data provider. The cost used by both the data provider and the data collector before the modification, was $2400 whereas after the modification the figure went down to $2200. Because of a lot of the responsibilities allocated to the data miner before the modification the mining cost was $25000 but after the modification the cost went to $15000. This is because of the removal of the identification of the business problem as a responsibility to the data miner.
The cost of decision making also reduced because of proper data mining and that reduced the time required for the analysis before decision making. The cost of decision making reduced from $11000 to $7000 because of the introduction of the system in the process.
The overal cost of data mining reduced from $38400 to $24200 which reduced by a good figure of $14200
Börger, E., 2012. Approaches to modeling business processes: a critical analysis of BPMN, workflow patterns and YAWL. Software & Systems Modeling, 11(3), pp.305-318.
Chinosi, M. and Trombetta, A., 2012. BPMN: An introduction to the standard. Computer Standards & Interfaces, 34(1), pp.124-134.
Geamba?u, C.V., 2012. BPMN vs UML activity diagram for business process modeling. Accounting and Management Information Systems, 11(4), pp.637-651.
Han, J., Pei, J. and Kamber, M., 2011. Data mining: concepts and techniques. Elsevier.
Zur Muehlen, M. and Recker, J., 2013. How much language is enough? Theoretical and practical use of the business process modeling notation. In Seminal Contributions to Information Systems Engineering (pp. 429-443). Springer, Berlin, Heidelberg.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
