Integrating Agile Planning and Estimation into Traditional Methods for Large Projects

Agile processes are often considered challenging to adopt in larger and formal project development and management environments. It may not be possible and need for an organisation to adopt a pure agile process for a large project. How can organizations adopt agile planning and estimation best practices within their formal and generic process lifecycle (GPL)? This article presents an agile planning and estimation model that can be used for large projects as a guide.

Agile Process

The core of the agile process is a collection of sprints or small increments. The agile process can be classified into three key categories or stages:

·       Pre-Sprint – focus on the analysis activities of the GPL

·       Sprint – focus on the planning and development activities of the GPL

·       Post-Sprint – focus on the sprint post-mortem and release activities of the GPL. 

The agile process has practices that can be executed within a GPL to produce the specific project deliverables. The following are the key practices, which have been grouped in different process stages: 

·       Pre-Sprint - focus on the analysis activities of the GPL

o   Capture business requirements or user stories

o   Develop solution architecture

o   Identify system uses cases supporting business requirements

·       Sprint – focus on the planning and development activities of the GPL

o   Plan for the sprint

o   Detail use cases for the sprint

o   Design and develop (code and unit test)

o   Develop integration testing cases for the sprint

o   Conduct code reviews for the sprint

o   Conduct user viewings for the sprint

o   Perform integration testing for the sprint

o   Manage requirements changes

·       Post-Sprint – focus on the sprint post-mortem and release activities on the GPL

o   Conduct sprint review and retrospective

o   Release sprint for UAT

 Agile Planning and Estimation

The following model presents the three estimation levels in relation to agile planning and estimation for the large projects.

Figure 1: Agile Planning and Estimation Model

Project team reviews the project brief (pre-sprint) and provides Level 1 initial estimates based with some percentage variance (e.g. +/- 50%) for the project in hand. Project team can provide more accurate Level 2 estimates with some percentage variance (e.g. +/- 10%) based on the more detailed information captured via business requirements (pre-sprint). Further, project team can provide more accurate Level 3 estimates with some percentage variance (e.g. +/- 5%) for each individual system use case listed in the sprint for sprint planning (sprint). Level 3 estimates can be adjusted throughout the sprint and also reflect any corresponding changes in the level 2 estimates accordingly for the real time project progress (e.g. increase or decrease in estimates). Project team can record actual figures in the end of each sprint and reflect any changes in the level 2 estimates accordingly for the real time project progress (post-sprint).

The following sprint and project burn charts can assist with the real time tracking of the estimation and progress.

Cost Burn Chart: Level 2 amount on y-axis and sprints on x-axis showing relationship between the sprints the amount consumed to date.

Days Burn Chart: Level 2 Days on y-axis and sprints on x-axis showing relationship between the sprints and the number of days consumed so to date.

Use Case Burn Chart: Number of project uses cases on y-axis and sprints on x-axis showing relationship between the sprints and the number of working use cases that have been completed (released for UAT).

 

Comments

  1. Business Transformation Strategy is change management strategy to get more and more business leads to bring business on the peak of market with the digital strategies.

    ReplyDelete

Post a Comment

Popular posts from this blog

What is Information? A Theory of Information Trilogy

Circular Information Lifecycle Management for Enabling Circular Economy

How to define information or data strategy?