#WHY IS BUSINESS PROCESS MODELLING IMPORTANT SERIES#
The team at HEFLO defines business process modeling as “a set of activities that must be followed to allow the creation of one or more models for representation, communication, analysis, design, synthesis, decision making and control of business.”Įssentially, business process modeling is any action or series of actions taken by business analysts, company managers and the C-Suite to improve business operations, making them more efficient.īusiness process modeling has become an increasingly popular tactic in the corporate world. It visually describes a process from start to finish, so anyone can see how your company operates. What Is Business Process Modeling?īusiness process modeling is meant to map out how your team, department or business gets work done. You don’t have to be an expert BA to try business process modeling, but you do need a drive to improve your department’s efficiency and help your company. This strategy is used in small companies of a dozen people up to major corporations. Whereas in the model monitoring stage, we focus both on the statistical as well the business metrics to derive our conclusion of being confident in the relevance and the reliability of a particular model.Have you ever stopped to ask why you do a specific task each day at work, and whether it can be done more efficiently? Do projects or plans get bogged down to the point where they are late and consistently over budget? If so, business process modeling can help.īPM is a tool that analysts and project managers use to identify weaknesses and redundancies in companies. In the model validation stage, we focus majorly on the statistical metrics that can decode the model performance and response for us. This is the most important distinction between the two stages. Once the model is deployed, the model monitoring processes ensure the relevance of the model by judging the population distribution and also recording back-dated error % comparisons between the model predictions and actuals data as soon as that starts coming in to make sure that the model performance is in the acceptable range.
The purpose of model validation is to check the accuracy and performance of the model basis on the past data for which we already have actuals.
Also, we check that the population distribution should not be significantly different as compared to the development time period to ensure that the model is still relevant and okay to be used. A specific monitoring frequency is decided for every model and it’s evaluated then to make sure that the model is performing up to the mark and its results are reliable. Model Monitoring comes into effect after a model has gone into the PROD (production) stage. This is immediately after the model development. If a model fails to perform in the validation stage, it goes back to the development stage. Model validation is carried out in tandem with the model development process. To begin understanding this distinction, we’ll cover three-pointers: The purpose, the metrics, and the time period.