Something like: "I needed to understand a process flow, so I used brainstorming, then interrelationship diagraph, then made a flowchart, and finished with a SIPOC to see context." In your reply, can you name your objective, then list the tools you used in order? That way we may begin to see how tools can flow into a solution.
If you want, you can elaborate on how you linked the tools into a series. If you used any of these tools in an unusual way to make the series reach your objective, that would be something great to add.
To get started, here is a list of common tools:
- Mind mapping
- Fishbone diagrams (or Ishikawa if you prefer)
- Tally sheet (usually called check sheets in quality discussions)
- Is- Is Not Matrix
- Run charts (separate from control charts, I think, as those are a system of themselves)
- Five Whys
- Flowcharts (basic or various types)
- Relationship diagram (or interrelationship digraph)
- Affinity diagram
- Tree diagrams
- Pareto charts
- Scatter diagrams (X-Y plots)
- Work Breakdown Structure
- Stakeholder analysis
- Matrix diagrams
- Process decision program charts
- Gantt chart
- Activity network diagrams (or arrow diagrams)
It seems that I only use tools in combinations. Years ago I noted that each issue analysis or problem-solving event has three stages that often require three different types of methods or tools. An opening event requires methods that increase the number of options that are considered (e.g., tools like mind-mapping, heuristic analysis, SCAMPER, and various types of brainstorming (e.g., TRIZ). In the second stage the methods need to focus on sorting by eliminating duplicates, establishing relationships (such as interactions, correlations, or subservient structure (i.e. subset decompositions). Here methods like tree diagrams, KJ (affinity) analysis, risk trees (e.g., probability trees for failure analysis decomposition like Fault Tree Analysis), and morphological box can all apply. The final stage is closing where the solution state becomes the focus among all alternatives. My choices here tend toward management by fact and data-driven decisions using Individuals Control Charts (with tests 1 and 2 only), Cp/Cpk process capability studies, ANOVA, Pareto charts of main effects magnitudes, and Yamaguchi Charts (e.g., stacked bar charts by process step which illustrate the three types of work (value-added, required NVA and pure NVA). These methods are then supplemented by RCA.
I must admit that most often the issues and problems that I face are the more complex ones that organizations have difficulties in resolving by themselves. However, the three-phased pattern (opening, sorting, and closing) can be applied with a variety of methods and tools across a variety of problem complexities.
Thank you for the points you make. I have intuitively felt these 3 stages apply, but you put it into words very clearly. I am curious as to why quality communities have focused so much on individual tools and buzzword-oriented tool sets like six sigma and Lean, and not on the process of improvement in a more general sense, using this 3 stage paradigm (or something like it) and mapping out common 'tool strings'? It would be like having a series of standard processes for improvement that could use a guide (undeveloped as of yet) to choose how to approach an issue. My point here is seek some of those 'tool strings' that others have found useful.
I like your three stage model for problem-solving and I think I have intuitively approached most major problems I have encountered in exactly this way. I like how you loosely categorized the tools into the three phases and this might help me understand when to use them in the future.
I share your curiosity about how people use tools together to tackle difficult issues. As I have learned about quality tools, I have often struggled with what context the tools would be used in. Especially if they are significantly different from the tools I most often use.
Thanks for the great topic!
I have a perspective on this "singularity of tool selection" approach. Ishikawa and Kume both presented sequences of tools to be used by QC Circles and for continual improvement (kaizen) efforts. However, when I joined the throngs of many Americans in the study of Japan after Deming's 1981 NBC White paper, I was part of a HP team that was attempting to find the "silver bullet" that enabled Japanese semiconductor and electronics industries to advance (and surpass) their American competitors. We found many tools; however, it took almost 5 years for us to realize that they existed in integrated tool-sets - combinations of methods and tools applied to address individual issues. One illustration from a Toyota visit might help. We have all heard of the eight wastes; however, on a visit I noted that Toyota had a set of standard approaches (e.g., toolkits) that were used to attack each waste. We understood them as singularities while Toyota saw them as an analytical process tied to each of the individual wastes. Many of these tools or methods were "reused" in several of the "waste management toolkits" (as I thought of them): 5 Why's, 3-S, SMED, Poka-Yoke, and Kanban were all popular). However, when teaching a child something new you start with "baby steps" not by learning how to sprint. The problem, as I see it is that all too often we never learned how to become a world class sprinter. This becomes even more problematic when we come to the digital age where simultaneous data collection from multiple sensor points of view provide us with the need for solving multiple, simultaneous equations in multiple variables - a vector space solution, while we have been trained in a single-variable contained space. We need to learn to live in a stochastic process in multi-dimensional space which extends across time boundaries. Is our community ready to learn how to walk, then jog, then run, then sprint?
Your post brings to mind a few things.
First of all, it took 5 years to realize the use of integrated tool sets!!!! I suppose this idea was so culturally normal, for the Japanese, that they did not emphasize it in conversation.
Second, we in the west tend to focus on simple 'silver bullets' like "six sigma" or "lean" that are very large tool sets and approaches. They are treated as one-size-fits-all tools sets.
Third, your post moves into the territory of cultural management, as befits your standing as a leadership coach. There is indeed a missing element in our teaching of process management where we try to leap from single tools to massive reengineering approaches. I think there is a need to change leadership coaching to 'think medium' and emphasize making standard toolkits for specific issues.
This is important, and looking at our CMQ/OE BoK we do not cover this as the BoK is based on current practices, not on needed future practices. I think QMD as well as the other divisions are the right places to make a 'future BoK' as a supplement to today's BoK. The concept of a 'future BoK' belongs in our QMD Leadership are, I think. Perhaps it belongs in a society-wide discussion area?
My intent in this particular post is to try to find current usage of these 'tool sets'. What are they? Where are they used? How are they used? I am thinking of making a series of learning videos for sale in my business, and I am trying to use MyASQ as a research tool for this. While at it, this may become fodder for a new book or two, most likely published by ASQ Press to the benefit of the society. I am switching hats here, from my business hat to my volunteer hat and back again. As I have said before, I have two bad habits, eating regularly and sleeping indoors. My business venture helps me keep my bad habits.
Once the Value Stream is mapped, then a list of opportunities for improvement is generated: brainstorming
One next step might be to categorize types of opportunities: affinity chart
Or another might be doing the pain index: prioritization and Pareto or prioritization matrix
Once the highest priority cycle time reduction point is identified, Gemba to gather specific data on the bottleneck
Exploit the bottleneck: Root Cause Analysis
And on it goes.
This is only the sequence I have used. Others have different perspectives and would choose different tools. We gravitate to those that match our problem solving approaches.
Thank you for your 'string' If we can find out what others do, we may find more commonality than you expect.
2. Key Observations/Data Collection [P]
3. Data Analysis> Problem Statement [P]
4. Ishikawa Diagram [P]
5. 3 Legged 5 Why [P]
6. Just Do It Change (Minor issue; solved by 5S etc.) [D]; or
Process Improvement (Process issue) [D]; or
Kaizen (Systemic issue) [D]
7. Verification of Effectiveness Follow up (C)
8. If effectiveness is verified make process/system change permanent. Update documentation [A]
9. Lessons Learned are then incorporated into Work Instructions, Troubleshooting Guides and/or PFMEA as deemed appropriate. [A]
1. Data collection (Part Number, Date Code, Test Data (located in a 2D Laser Etched Barcode), Qty of Parts affected. Location in skid of affected parts, Picture of skid, Picture of Parts, Description of issue
2. Determine if parts should be Returned, Sorted in Place or can be Deviated.
If Need to be Returned,
3. Issue Return Authorization.
4. Sort stock (If a build ahead part); If built to order, then generate a Quality Alert, have operators sign off after they have been made aware and layer audit to verify that the Quality Alert is being followed.
5. After Sort or Built to Order activity certify parts (Digital version of the Certification Document needs to be emailed to the Customer and placed on the skids.)
6. Ship Certified product to the Customer.
7. Once parts are returned start problem-solving methodology.
8. Inspect reworked or conforming component parts.
If Sorted in Place,
3. Issue sort parameters and rework instructions, if appropriate, to Internal Personnel (Travel permission & arrangements needed.), Outside Sorting Firm (Payment set up needed.) or Customer personnel (Labor Payment and line shutdown fee, if necessary.)
4. Execute sort at Customer.
5. Scrap or Return parts to plant, as appropriate.
6. Once parts are returned start problem-solving methodology.
7. Customer uses remaining conforming parts.
3. Record Deviation in system.
4. Issue Request Authorization for 1 representative part.
5. Once part is returned start problem-solving methodology.
6. Customer uses remaining parts.
These are great! Can you list some of the quality tools you use to accomplish these activities?
- After the initial data collection. (Part Number, Date Code, Test Data (located in a 2D Laser Etched Barcode), Qty of Parts affected. Location in skid of affected parts, Picture of skid, Picture of Parts, Description of issue). We will bring the part(s) back in to evaluate.
- Once received we have a checklist that we go through for every product that is returned. This would include standard functionality and critical dimensions.
- Then we look at the actual complaint and a common issues checklist that has been derived from lessons learned.
- If the issue matches a common issue on that checklist, then we audit the process to look at where it broke down via a process flowchart, workflow diagram & PFMEA.(Brainstorming, needed.)
- Then we look at why it broke down utilizing Ishikawa Diagram & 3 Legged 5 Why. (The 3 Legged 5 Why is completed at the end when we have a prevention measure that has been verified.)
*[If the root cause is elusive or there are too many variables we incorporate brainstorming, affinity diagram or scatter diagram. Sometimes we even need to incorporate new tools on a temporary basis unless the prove to be superior to what we are using otherwise. Any new issues are lessons learned and incorporated into Troubleshooting Checklists (guides), PFMEA, work flows & flowcharts as appropriate.]
Tools used for step 6-9: A solution could be as easy as a 5S assessment with the specific problem in mind, this would be a Just Do It improvement activity. Then testing the change to verify effectiveness. A process improvement would include a process capability study, an updated process flowchart, updated work flow, updated PFMEA, updated troubleshooting checklist (guide). A systemic issue we would use whatever most of the tools above as well as value stream mapping for current state and later a future state version during the kaizen event. All of these improvements would first be tested to verify effectiveness prior to rolling them out as official. Also, all done utilizing PDCA.