Open Universiteit

Please use this identifier to cite or link to this item:
Title: Specification, Modeling and Validation of KPIs and their Required Properties
Authors: Scholten, C.F.
Keywords: requirements
performance measures
performance indicators
key performance indicators (KPIs)
Issue Date: 23-Aug-2017
Publisher: Open Universiteit Nederland
Abstract: A ‘Key Performance Indicator’ (KPI) is defined as “a metric used to quantify the efficiency and/or effectiveness of an action that is compared to a given norm or target which is derived from business strategy”. KPIs should be specified, modeled and validated to have specific properties. These required properties are: quantifiability, sensitivity, congruency, reliability, efficiency and improvement-orientation. But with standard and popular modeling notations, such as those in ‘Unified Modeling Language’ (UML), the properties cannot be guaranteed. In this work, recommendations on properties are combined with a constructive definition into a conceptual model of KPIs and their required properties. This model can be used for the specification, modeling and validation of KPIs. Also, the modeling semantics that are needed to support the modeling and validation of KPIs are explained. The modeling notation that is used should not only include semantics of business objects and their attributes, but also of states and events that are extended with data to make up transitions in the object-lifecycles, and CSP-composition of the transitions to compose executable and deterministic behavior. The semantics are included in the method EXTREME which uses ideas from approaches to goal modeling and Protocol Modeling.As a case study, the time-related KPI ‘Percentage of In-Time Processed Orders’ of business process ‘Order to Pay’ is specified, modeled and validated using the conceptual model and a protocol model. The KPI is initially specified as the percentage of closed orders for which the ‘Processing Lead Time’ equals or is lower than the ‘Norm Lead Time’. Using the conceptual model this is further specified. By developing and executing a protocol model, it is validated that the KPI has its value calculated from the right number of instances of business concepts (quantifiability); changes in states and attributes-values of the instances (sensitivity); a clear mathematical relationship (congruency); and a minimum number of assumptions (reliability) and scenarios for manipulation (improvement-orientation). Future research is needed on the risk that KPIs in general or in specific (time-related) cases might be used to manipulate and/or do not reflect real performance or improvement.
Appears in Collections:MSc Business Process Management and IT

Files in This Item:
File Description SizeFormat 
AF-Final_Version_v2.2_C.F.Scholten.pdf1.04 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons