3 min read
Question: What is continuous improvement?
Short answer: a structured approach using tools or methodologies to systematically improve business outcomes.
Longer answer: Whether the problem is cost, schedule, or throughput related, Speed to Solution’s Continuous Improvement approach can not only help isolate the root cause of the problem, but help to lessen the pain. Unlike some die-hard Six Sigma statisticians who express disdain for that soft lean stuff or some of those Lean zealots who refuse to do math, we embrace any tool that helps solve a business problem. Thus, we have spent decades training ourselves and out team in Lean, Six Sigma, Theories of Constraints, Systems Thinking, and Data Science techniques so we can bring the right approach to your challenge.
We will touch on each of these specific approaches in more detail shortly (no, they are not simply executive buzzword bingo terms), but let us start with something fundamental to problem solving: the scientific method.
- Observe
- Hypothesize
- Test
- Analyze
- Update mental model and repeat
This basic approach forms the foundation of people’s desire to better understand the world around us. As I write this, I have a 4 year old daughter, who intuitively understands this process… sometimes better than a Vice President! In our years of Continuous Improvement practice, too often we see managers and executives lose sight of this approach in favor of politics or short cuts. To quote J.R.R. Tolkien in The Fellowship of the Ring: “Short cuts make long delays.”
We strive to unambiguously understand your challenge and apply this process, following the people, process, and data to provide improved results for your business.
With that introduction, let us look at some of the most popular continuous improvement methodologies so we can better understand their strengths and weaknesses.
Lean
Lean is a philosophy that embraces the tacit knowledge of the workforce to incrementally get a process closer to perfection. Practitioners typically follow a Plan, Do, Check, Act cycle (sound anything like the scientific method?). Tools include detailed process observations, process mapping, rapid improvement (or Kaizen) events, an iterative method to signaling when action should occur (Kanban), and waste (muda) identification and removal. If you cannot tell from the parentheticals, Lean was largely derived from the Toyota Production System.
By the way, do not be convinced of any mystical power of these terms. Lean comes largely from the Toyota Production System. Toyota is based in Japan. In Japan, they speak Japanese. These are Japanese words for useful concepts to make processes work better for your customers and ultimately for your business.
Six Sigma
Six Sigma is a data-driven approach to understanding and controlling variation to obtain a consistent, desired process outcome. Practitioners often follow a Define, Measure, Analyze, Improve, Control cycle (again, sounds a lot like the Scientific Method). Much of Six Sigma relies on gathering data with the intent to precisely show the root causes of process variation. Once the vital few root causes are identified, teams work to eliminate them at the source, resulting in much more consistent results. Six Sigma tools include graphical analysis, correlation/ regression, ANOVA, hypothesis testing, and fault trees.
Six Sigma offers a very robust approach to understanding how a process behaves. In a data light environment, it can take time to gather sufficient data to make a fully informed decision.
Theory of Constraints
To quote Goldratt, the Theory of Constraints’ creator, himself, this approach is all about “focus.” Where Lean and Six Sigma attempt end to end improvements, the Theory of Constraints suggests there is a single bottleneck impeding a process’ ability at any one time. If you find that pressure point and alleviate it, the process immediately gains ability up to the next bottleneck. The implications of this are pretty amazing: with far fewer resources, one can achieve maximum improvements immediately.
Goldratt also espoused a structured approach to process improvement: identify the constraint, subordinate all other processes to the constraint, elevate the constraint, and then identify the new constraint. Tools used by practitioners in this domain include work in process analysis, Little’s law, line of balance, the drum-buffer-rope, and queueing theory.
Systems Thinking
This approach is often the least familiar of many traditional process improvement approaches. And with good reason: it comes from system engineering rather than process. (Michael Hammer’s process re-engineering is probably the closest). With system thinking, our aim is to explicitly define and understand the top level (system) requirements rather than improve any specific process that may or may not achieve the ultimate end goal.
This philosophy is best summed up as solving a problem is not the same as getting what you want. We strive to step back from the immediate task and think more broadly about the organization’s objectives and how the current system supports those needs. Where gaps exist, decomposition and allocation allows for identifying improvement opportunities and matching solutions.
Data Science
Data analysis fuses the domains of business, statistics, and computer science together. As a discipline, it frequently seeks to inform the decisions leadership needs to make. The international standard process for data science is CRISP-DM:
- Business Understanding
- Data Understanding
- Data Preparation
- Modelling
- Evaluation
- Deployment
We do our best to work with the data that you have. We do not require a Hadoop cluster and an army of Python speaking PhDs to offer insight. There can be a lot of value from even a single data point under the right circumstances. To quote Charles Kettering, “A problem well stated is half solved”.
Continuous Improvement Wrap Up
This is a lightning speed tour of some of the most useful problem solving methodologies. As continuous improvement practitioners, we strive to use the right tool for the job rather than be acolytes for any single approach. We find that leading with the problem at hand rather than the toolset results in better business outcomes. Let us know if we can serve your organization.