While Six Sigma was initially a quality control system, it has also been very succesful in managing projects. The various stages of an analytical project can be divided using the DMAIC methodology.
DMAIC stands for
- Define
- Measure
- Analyze
- Improve
- Control
Related to this is DMADV, ( “Design For Six Sigma”)
- Define
- Measure and identify CTQs
- Analyze
- Design
- Verify
2) CRISP
CRISP-DM stands for Cross Industry Standard Process for Data Mining
CRISP-DM breaks the process of data mining into six major phases- and these can be used for business analytics projects as well.
- Business Understanding
- Data Understanding
- Data Preparation
- Modeling
- Evaluation
- Deployment
3) SEMMA
SEMMA stands for
- sample
- explore
- modify
- model
- assess
4) ISO 9001
ISO 9001 is a certification as well as a philosophy for making a Quality Management System to measure , reduce and eliminate error and customer complaints. Any customer complaint or followup has to be treated as an error, logged, and investigated for control.
5) LEAN
LEAN is a philosophy to eliminate Wastage in a process. Applying LEAN principles to analytics projects helps a lot in eliminating project bottlenecks, technology compatibility issues and data quality resolution. I think LEAN would be great in data quality issues, and IT infrastructure design because that is where the maximum waste is observed in analytics projects.
6) Demings Plan Do Check Act cycle.
DECISION STATS 