Muda Be Gone! Value Stream Management Reduces DevOps Waste

Today organizations operate faster and faster to develop quality software for their customers, adopting DevOps with the goal of continuously delivering value to users and back to the company. These organizations have development teams working all over the world using different tools, so having visibility into how value flows through the process is important. It is critical to use value stream management to gain insight into the bottlenecks and blockages that can lead to increased waste, and preventing value from flowing to the end user.

The eight forms of muda, or waste that occurs in software value streams, are identified as: defects, over productions, wait times, non-value processing, transportation (approvals), inventory (excessive backlog), motion and eschewed talent. You can remember these with the acronym DOWNTIME.

Let’s review the eight forms of muda and how value stream management with CloudBees DevOptics can help:

  • Defects are of course a huge cause of waste in software development and delivery, with each bug implying additional effort in rework.

    • Defects and the remedial work to address those defects are easily tracked visually in real time across value streams. New defects discovered at any point in the value stream are immediately visible.

  • Overproduction in software is the act of building things that users don’t want or don’t need.

    • Product managers and executives can see all of the work flowing through the value stream at any point in time or historically, and ensure gold-plating is avoided.

  • Waiting waste in software value streams can have many sources – commonly this is in manual handoffs between teams and processes, or waiting for build, deployment or test resources to become available.

    • Manual handoffs between different teams and Jenkins jobs are readily apparent, and insight is provided into waiting times and waste caused by unavailability of underlying resources in the CD platform via the Run Insights capabilities.

  • Non-value add processing waste in software might be in the form of test sequencing that doesn’t place scenarios most likely to fail up front, or in any unnecessarily cumbersome processes that detract from development team productivity.

    • Having real time visibility of the end to end value stream, coupled with the insight provided by the DevOps metrics make it much easier to identify problem areas impacted by non value-add processing.

  • Transportation waste in software can include things like manual approvals, and non streamlined approval processes with too many levels.

    • In CloudBees DevOptics, you can easily see where work is stalled at each gate, preventing the value from flowing. This can be in the form of some multi-layered manual approval process that blocks work from moving to the next gate.

  • Inventory waste refers to excessive backlogs of features or non-critical fixes that never get worked on. As is any code not yet delivered to production. The greater the length of time that work sits in a partially done state, the greater the risk it impacts other work and not useful to users anymore.

    • With CloudBees DevOptics its really easy to see and track how long any piece of work across the value stream sits undelivered to production or in a given gate.

  • Motion waste refers to unnecessary human motion. This is less relevant in software development and delivery, but it can be used to describe waste that occurs with rapid and repeated change of ownership of work, in other words, who owns what work, when.

    • CloudBees DevOptics shed light onto work items and tickets and their ownership, to help track waste created by motion of ownership.

  • Eschewed talent waste in software can be in numerous forms including any environmental or human factors that might impact development team productivity. Or the use of hard to use tooling. Or the waste created by too much task switching. Or the waste incurred by having visible metrics that are confusing to the team, or that encourage behavior that is often counter to the goals of value stream optimization. And of course any processes or methodologies that discourage developer creativity create waste in this category.

    • Eschewed talent waste is actively avoided by providing teams with CD metrics that are key leading indicators of organizational performance, focusing improvement efforts on the right things.

The CloudBees suite of CloudBees Core and CloudBees DevOptics makes value stream management easy. The solution provides a continuous integration/continuous delivery (CI/CD) automation engine with the flexibility to support diverse software portfolios and the unified governance required by growing organizations. Based on Jenkins, CloudBees Core extends Jenkins with functionality that embeds best practices, rapid onboarding, security and compliance. It provides a shared, centrally-managed CI/CD service but delivers a self-service experience for engineering teams.

Get CloudBees DevOptics free today to help you gain insight into your software delivery value streams.

Additional resources