CSDV3017  ·  DEVOPS  ·  SCHOOL OF COMPUTER SCIENCE, UPES

DevOps Adoption,
Lean & Kaizen Principles

Lecture 7 — Unit III opens: how organisations actually move to DevOps, the manufacturing philosophy it borrows from, and why adoption is harder than installing a pipeline.

InstructorDr. Mohsin Furkh Dar
SessionWeek 3 · Mon, 22 Jun 2026
Time14:00 – 15:00
UnitUnit III
I
II
III
IV
V
VI
VII
WHERE WE LEFT OFF

Recap & today's agenda

Lectures 1–6 · Done

Unit I & Unit II

  • What DevOps is, how it works, its benefits
  • Agile vs DevOps, Kanban & Scrum
  • DevOps tooling: CI, CD, monitoring, DevSecOps, SRE
  • Business case for DevOps; cross-functional teams
  • CI vs CD vs Continuous Delivery; tools for Agile
  • Silos as the enemy of flow — case study
Lecture 7 · Today

DevOps Adoption, Lean & Kaizen

  • What "DevOps adoption" really means
  • A maturity path: crawl → walk → run
  • Lean thinking and the Toyota Production System
  • The 7 wastes — mapped to software delivery
  • Kaizen: continuous, incremental improvement
  • Real challenges teams hit when adopting DevOps
THE GAP

Tools were never the hard part

"You can buy a CI server in an afternoon. You cannot buy a culture that trusts a green pipeline over a manual sign-off." — the adoption problem, in one line

Unit I/II gave us

The "what"

Practices, tools, and the theoretical case for DevOps.

Unit III gives us

The "how"

How an organisation actually changes its habits to get there.

Industry reality

The "why it stalls"

Most failed DevOps initiatives fail on people and process, not technology.

DEFINITION

What is "DevOps adoption"?

DevOps adoption is the organisational journey of shifting how people, process, and technology work together — moving from siloed, hand-off-driven delivery to a shared, automated, feedback-driven flow from idea to production.

People

Culture & mindset

Shared ownership, blameless learning, breaking the Dev-vs-Ops mentality.

Process

How work flows

Smaller batches, faster feedback loops, work made visible end-to-end.

Technology

Enabling automation

CI/CD, IaC and monitoring — but only as the enabler, never the goal itself.

HOW ADOPTION HAPPENS

A maturity path, not a switch

DevOps is never "turned on" in one release. Most organisations move through recognisable stages — each one building trust before the next is attempted.

Stage 1 · Crawl

Isolated wins

One team automates its own builds and tests. Proves the idea on a small, low-risk surface.

Stage 2 · Walk

Cross-team flow

Pipelines connect Dev → QA → Ops. Shared dashboards replace status-update meetings.

Stage 3 · Run

Org-wide capability

DevOps becomes "how we work" — self-service infra, continuous deployment, shared on-call.

WHERE LEAN COMES FROM

Lean: born on the factory floor

The Toyota Production System

Lean thinking originates in 1950s Toyota manufacturing: the goal was to deliver exactly what the customer needs, in the right amount, at the right time — with the least possible waste.

DevOps borrows this thinking wholesale: a deployment pipeline is a production line for software.

GOAL: Highest Quality · Lowest Cost · Shortest Lead Time
Pillar
Just-In-Time
build only what's needed, when needed
Pillar
Jidoka
automation with a human touch — stop on defects
FOUNDATION — Standardised Work · Kaizen · Heijunka (levelled flow)
LEAN VOCABULARY

The 7 wastes (Muda) — in software terms

01

Partially done work

Code sitting in branches, unmerged, untested.

02

Extra processes

Redundant approvals, change boards, duplicate sign-offs.

03

Extra features

Gold-plating — building what nobody asked for.

04

Task switching

Engineers split across too many projects at once.

05

Waiting

For environments, approvals, test results, releases.

06

Motion/handoffs

Work passed Dev → QA → Ops with context lost each time.

07

Defects

Bugs found late, costing far more to fix than early.

Spotting these in your own delivery process is the first concrete Lean skill — and the starting point for any Kaizen improvement.

FROM FACTORY TO PIPELINE

Lean thinking applied to DevOps

Value Stream Mapping

See the whole flow

Chart every step from commit to production, and mark where work waits versus where it's actually worked on.

Flow

Small batches win

Smaller, more frequent releases reduce risk and surface problems while they're still cheap to fix.

Pull, not push

Work is pulled, not pushed

Teams pull the next item when they have capacity — mirrored directly by Kanban boards and WIP limits.

改善 KAIZEN

Kaizen: change for the better

The core idea

Kaizen means continuous, incremental improvement — made by the people doing the work, every day, rather than occasional big-bang transformation projects.

It assumes no process is ever finished; it can always get a little better tomorrow.

Two flavours

  • Flow Kaizen — improving the value stream as a whole (the manager's lens)
  • Process Kaizen — improving how a specific task is done (the team's lens)

DevOps relies heavily on Process Kaizen: engineers closest to the pipeline fix friction first.

SEEN IN THE WILD

Kaizen, in everyday DevOps practice

Daily

Retrospectives

Short, regular reflection on what slowed the team down — and one small fix to try next.

When things break

Blameless postmortems

Failures are treated as data about the system, not as someone's mistake to punish.

On the pipeline

Andon-style stop-the-line

Anyone can halt a broken build — the same principle as a factory worker pulling the andon cord.

Notice the pattern: Lean gives DevOps its map (where the waste is), Kaizen gives it its habit (always closing the gap).

TYING IT TOGETHER

CALMS — where Lean meets DevOps culture

C

Culture

Shared ownership over hand-offs and blame.

A

Automation

Removing manual, repeatable toil from the pipeline.

L

Lean

Small batches, visible flow, waste eliminated.

M

Measurement

Lead time, deployment frequency, change failure rate.

S

Sharing

Knowledge and incidents shared openly across teams.

Lean & Kaizen sit at the core of two of these five letters.

WHY ADOPTION STALLS

Challenges of DevOps adoption

People

Cultural resistance

Teams comfortable with existing roles resist shared ownership and accountability.

Structure

Organisational silos

Dev, QA and Ops report to different leaders with different, sometimes conflicting, incentives.

Tooling

Tool sprawl

Adopting tools without aligning process first creates automation around a broken workflow.

Skills

Skills gap

Engineers need new scripting, cloud and automation skills that take real time to build.

Legacy

Legacy systems

Monolithic, tightly coupled applications resist fast, independent, frequent releases.

Leadership

Measuring success

Without the right metrics, leadership can't tell real progress from busywork.

QUICK REFLECTION

A familiar adoption story

Class discussion

The "silo case study" from Unit II — revisited

Recall the silo case study from Lecture 6. Now reframe it through today's lens: which of the 7 wastes were visible in that scenario? Was the fix mainly cultural, process-based, or technical — and which Kaizen habit would have caught the problem earliest?

01
Identify the waste
02
Name the blocked hand-off
03
Propose one Kaizen-sized fix
WRAP-UP

Today, in three lines

  • DevOps adoption is an organisational journey across people, process and technology — not a tool purchase.
  • Lean gives us the vocabulary to find waste; Kaizen gives us the daily habit of removing it.
  • Most adoption failures trace back to culture, silos and unclear metrics — not to missing tools.
Next lecture · Lecture 8

Tue, 23 Jun 2026 · 12:00–13:00

Monolithic vs Microservices architecture; Cloud & DevOps resiliency.

Before next class

Quick prep

Think of one waste from today's list that you've personally experienced in a group project or internship.

CSDV3017 · DEVOPS
SHEET 01/14