Business Process Modeling life Cycle
In this article, the challenges and risks involved in Business Process Re-engineering within the problem domain of a large retail bank are discussed. We demonstrate how the tools of the ZDLC platform are employed to mitigate the risks and address the challenges of automating and transforming a manual intensive business process into a lean and productive business process. In lean we mean speed up the process whilst achieving low cost.
Cognizant presented an approach for the Business Transformation programme using industry standard DMADV (Define, Measure, Analyse, Design and Validate) of the DFSS (Design For Six Sigma) methodology. The latter has been enriched by utilising the tools of the ZDLC platform to accelerate the process of DMADV through intelligent automation to reduce manual effort in the software development life cycle. The proposed model to integrate ZDLC in a software solution delivery process is depicted as follows:
The principle focus areas of the software development lifecycle we aimed to improve are defined as follows:
- Requirement Elicitation – We proposed the use of the House of Quality – enhanced (HoQ-e) to gather and formulate requirements. HoQ-e ensures requirements are systematically refined in a reliable and traceable drill down process, i.e. from the high level business requirements to user requirements to technical requirements. It preserves the traceability throughout the lifecycle. Traceability is an essential capability to constantly validate the requirements against business goals.
- Requirement Validation – The HoQ-e tool is empowered with techniques that improves the method of validating requirements against business goals or business drivers. The HoQ-e tool of the ZDLC platform, is unlike the classical HoQ proposed by the Quality Function Deployment (QFD) suite. The HoQ-e has seamlessly integrated key techniques of the QFD and ensures the techniques are logically and non intrusively bound together. The techniques are: 1) AHP (Analytical Hierarchy Process) – allows pair-wise comparisons of requirement attributes to minimise the inconsistencies in the activity of subjective prioritisation; 2) Affinity Diagram – enables the grouping of requirements sharing common characteristics together which is a key exercise of abstraction for devising the architecture of the solution, 3) Value Stream Map – enables business decision makers to identify the parts of their business processes which provide the most value to their customers and markets and 4) GQ(I)M – (Goal Question Indicator Matrix) – derives measurable quality attributes, i.e. CTQs (Critical To Quality) from high level non-functional requirements, SLAs or inefficient drivers of the business process.
- Quality Modelling – The non-functional requirements of the solution are derived from the Inefficiencies Drivers, articulated by the business stakeholders and captured in the HoQ-e using the method of GQ(I)M. The latter warrants the formal testability of requirements and rich measurement criteria defined to ensure that requirements deliver desired business benefit.
- Design Validation - We proposed the use of the Testable Integration Architecture (TiA-e) tool to automate and reinforce the exercise of validating the design artefacts. TiA-e integrates the formal methods of pi-calculus and Coloured Petri Nets (CPN) to simulate and compile the design artefacts (in the context of this programme, the design artefacts were in the shape of BPMN 2.0) against the requirements (output of the HoQ-e) to check for conformance. As a result, TiA-e ensures quality is built into design.
- Automatic Root Cause Analysis – At the System Integration Test phase, we employed the Systemic Defect Profiler (SDP-e) tool of the ZDLC platform to automatically validate the implementation of the code (Run-time) against the expected design expectations (Design-time). It reduces the cost of fixing defects by minimising the effort required to find the root cause of a defect.