Integral Logistics Management — Operations Management and Supply Chain Management Within and Across Companies

18.2.5 PDCA Do Phase — DMAIC Improve Phase

Intended learning outcomes: Produce an overview on tools of the jidoka concept, such as Andon (or a visual control system in a job shop) and Poke yoke (or fail-safe techniques).



Both the Do phase of the PDCA cycle and the Improve phase of the DMAIC method can be regarded as the most creative part of quality management tasks; namely, solution finding. The task is to generate a number of possible solutions that counteract the root causes of variations and defects.

In the manufacture of physical products, the individual process steps are usually described in quite great detail. The process step instructions often include measures with tolerances that must be adhered to during machining. The same should hold for processes in the information flow of a company, for example, for order processing. Exact descriptions are just as necessary for service processes, even though here it can be considerably more difficult to determine targets and variations.

For this creative process, the following principles hold (these are realized, for example, in the jidoka concept; see definition in Section 6.1.1):

  • Possible defects should be identified at the source of origin as early as possible in the process. The problems or defects can be identified by the human eye or by specialized sensors.
  • All components and units should be checked 100% to ensure complete faultlessness.
  • Direct intervention prevents further subsequent mistakes. With jidoka, any worker can stop the line by pulling a cord; this is then signaled on visual display devices called Andon boards that are visible to all.
  • The processes must be made “foolproof” (poka yoke).
Andon is a visual signal, or visual control system. With jidoka, Andon are electronic display boards that show the status of the processes in a job shop or production line as well as information for coordination of the connected workstations. Commonly used colors to indicate status are green (OK), / yellow (needs attention), or / red (stop). See here [Toyo98].

Figure 18.2.5.1 shows an example of numbered Andon lights.

Fig. 18.2.5.1       Andon: visual control system in a job shop.

The yellow light shows a problem at workstation 1. There the worker has pulled the cord to stop the line.

Poka yoke, or fail-safe techniques, means to avoid (yokeru) inadvertent errors (poka). The basic principles of poka yoke advocate designing or developing tools, techniques, and processes to detect defects, thus relieving people of error-prone tasks such as repetitive monitoring of the same states or the checking of a great many details [Kogy90].

Some examples of such simple mechanisms and devices are:

  • Sensors that detect missing or defective components and shut down the process automatically.
  • Guide pins on parts that prevent assembly in the wrong orientation.
  • Construction of parts and corresponding clamping and mounting apparatuses that prevent backward insertions.

Applying poka yoke to information flows represents a considerable challenge. However, for the area of order processing, computer-supported computer programs that can monitor the completeness of information and the correct sequence of functions are conceivable. Indeed, computer programs have always had the task of recognizing and preventing all possible defective constellations of data. Workflow techniques are now used to ensure correct implementation of additional sequences of ordered tasks, even where several people are involved in the process. Because of the many unforeseeable influences on information processes, however, these techniques are likely to be of help only for simple and highly repetitive processes.

Both the Do phase of the PDCA cycle and the Improve phase of the DMAIC method require reviewing and revisiting the deliverables as follows:

  • Have sufficient solutions been generated?
  • Were the solutions tested, and has the best solution been selected based on these tests?
  • For the selected solution, were target processes and cost-benefit analysis drawn up?
  • Was a plan drawn up for introducing the selected solution?



Course section 18.2: Subsections and their intended learning outcomes

  • 18.2 Quality Management Tasks at the Operations Level

    Intended learning outcomes: Produce an overview on the Deming Cycle (PDCA Cycle) and the Shewhart Cycle as well as the Six Sigma Phases. Present the phases of quality planning, control, assurance, and activation of the Deming Cycle. Describe the Six-Sigma phases of define, measure, analyze, improve, and control. Differentiate between continual improvement and reengineering.

  • 18.2.1 The Deming Cycle (PDCA Cycle) and the Shewhart Cycle

    Intended learning outcomes: Produce an overview on The Shewhart cycle developed in statistical quality control. Present the Deming cycle. Describe quality management tasks in the Deming cycle.

  • 18.2.2 DMAIC — The Six Sigma Phases

    Intended learning outcomes: Present DMAIC, the Six Sigma phases. Describe the tasks in the Six Sigma phases. Differentiate between DMAIC, RDMAIC, DMAICT, and DMADV.

  • 18.2.3 Quality Planning: PDCA Plan Phase — DMAIC Define Phase

    Intended learning outcomes: Identify the cause of differences between stakeholders’ expectations and actual product or process characteristics. Explain quality function deployment – the house of Quality and 10 steps of implementation. Describe the SIPOC diagram and the CTQ matrix.