Case Study: Integrating TRIZ Into Six Sigma

	Bookmark This Page Email This Page Format for Printing Submit an Article Read More Articles
	Related Tools & Articles Six Sigma Leadership And Innovation Using TRIZ Seeing a Forest of Innovations Because of TRIZ
	Discussion Forum "These two processes require fundamentally conflicting behaviors...While most organizations should do both - the activities should be led by separate groups, to avoid the conflicts that occur when any person or team attempts both simultaneously." Contribute to this Discussion

By Michael S. Slocum and Amir H.M. Kermani

Abstract

Six Sigma operates based on the assumption that the solution to a problem is contained within the process under investigation. Six Sigma is very powerful when this assumption is true as it enables the practitioners to identify the transfer function (y = f (x)) and control the output. In many cases the solution to the problem is not to be found in the process and this inhibits the ability to identify the control variables. In this case, a methodology that can solve the problem outside of the process boundaries, such as the Theory of Inventive Problem Solving (TRIZ), is necessary. The successful integration of TRIZ with Six Sigma can overcome any limitations originating from lack of solution location in the process space.

Introduction to Six Sigma

Six Sigma is a structured problem solving methodology used on any repetitive process, procedure or transaction. Statisticians at Motorola developed this methodology in the 1980s. The basic premise of Six Sigma is that there is a cost to poor quality, because poor quality causes lost sales and lost business opportunities. Improving quality through the reduction of defects leads to greater customer satisfaction. The primary goal of Six Sigma is to improve customer satisfaction by reducing defects and solving defect reduction problems. Six Sigma is a global problem solving process and defects can be related to any aspect of customer satisfaction: high product quality, schedule adherence, cost minimization, etc. In other words, every process, procedure or product has an opportunity to be executed correctly. Any opportunity that occurs which does not meet customer requirements is called a "defect" so with the correct definition of the problem Six Sigma can be used as a powerful problem solving process in any area of industry. The Six Sigma methodology consists of five phases: define, measure, analysis, improve and control (DMAIC).

Define: The life blood of a Six Sigma initiative is: What is the problem?

• A gap between what is happening and what we want to happen
• Something that does not allow a goal to be achieved
• A mistake, an error, a defect, a missed deadline, a lost opportunity or an inefficient operation
• An opportunity

A goal of Six Sigma is the reduction of defects to solve the problems. Therefore, defining the right problem is critical to the successful application of the methodology. The main goal of this phase is to define the problem, the variable that affects the problem and the elements crucial to quality.

Measure: The goal of this phase is transfer the current problem to the function y = f (X1,X2,..Xn). "Y" measures an attribute, which is important to the problem such as time, cost, etc. Y is dependent on a set of input variables. Each input is known as an "X" variable. In this phase we must be able to quantify the output of the process and every factor that influences the realization of the desired Y.

Analysis: The current situation is analyzed to identify the root cause of the problem, find correlations between the variables and get closer to identifying which X variables might truly be vital to the Y response (critical input).

Improve: Combinations of different feasible "settings" for the X variables are selected and the outputs Y are measured for each combination, so that the best possible solution is achieved.

Control: Maintain the critical input variable in the modified range, which prevents the problem from occurring in the long-term.

Figure 1: DMAIC process

Introduction to TRIZ

TRIZ is the Russian acronym for "Theory of Inventive Problem Solving." The method was developed by Genrich Altshuller and his colleagues in the former Soviet Union and now is a problem solving process used internationally. TRIZ research began with the assumption that there are universal principles of invention that are the basis for creative innovations for problem solving, and that if these principles were identified and codified, they could be taught to make the process of invention more predictable. The research over the last 50 years has found:

1. 1. Problems and solutions were repeated across industries and sciences.
2. 2. Patterns of technical evolution were repeated across industries and sciences.
3. 3. Innovations used scientific effects outside the field where they were developed.

TRIZ operates on a abstraction process that transfers the problem from specific to general or to a higher level of abstraction then was tried to solve the general problem. Using this process shows the universal principle for a group of problems. (See Figure 2.)

Figure 2: Abstraction Method

The mail goal of TRIZ is to transfer a specific problem to a general problem (problem formulation), then apply the universal principle to it and develop the universal principle in the specific area in which the problem occurs.

TRIZ Tools

TRIZ has a set of principles that may be used to resolve the contradiction in the system, based on observations collected from the analysis and decomposition of patents. This analysis allows the practitioner to convert, by abstraction, from the specific to the generic and allows the problem solver to take advantage of the fact that elements of the problem in question have already been solved by someone else.

Contradiction theory is the foundation of TRIZ. The abstraction of a generic problem may be carried out by identifying the technical contradiction in the system through the application of the 39 parameters or the establishment of a bi-polarity (A and anti-A). The user may then apply the 40 inventive principles to assist the creation of solution concepts. This process is conducted according to two four-step algorithms. The identification of the contradiction in the system (in some cases there may be several) is the root of problem solving using TRIZ. If the contradiction is hidden, or confounded, TRIZ-based modeling techniques may be used to identify the contradiction.

Overcoming Lack of Solution Location

The ultimate outcome of the DMAIC process is the optimization of process performance yielding an output asymptotically approaching entitlement. The application of the phases and phase specific tools is coordinated to produce this result. Design for Six Sigma (DFSS) can assist the creation of a new process that has an entitlement greater than the previous system. The DFSS System (DMADV) is based on the reduction of a concept to practice through the utilization of advanced tools for capturing and preserving the voice of the customer, integrating the functional capabilities of all responsible groups, and applying statistics and modeling to mitigate risk. However, there is no tool for the generation of the idea the DFSS process supports. This is the natural location for the integration of TRIZ and DFSS.

Use TRIZ in Each Phase of DMAIC

TRIZ may be integrated into each phase of DMAIC. By analyzing the output requirements for each phase you can capture a list of necessary skills. Mapping those required skills to the tools available in Six Sigma identifies gaps – TRIZ is the method that supports many of these gaps.

Case Study – Six Sigma and TRIZ Together

Define Phase: The Self-Heating Container

OnTech is the world leader in self-heating container technology, developing self-heating custom containers utilizing simple and cost-effective measures. The concept and construction is simple:

Figure 3: Self-heating Container Technology

Calcium oxide and water are used so that the energy released by an exothermic reaction heats the beverage. All the consumer needs to do is follow the simple instructions and wait approximately six minutes for a hot beverage or soup.

Measure Phase (With TRIZ)

Much work in the seaming of a metal end to a plastic body was performed by American National Can (ANC) on the Omni-Bowl Project. The double-seams achieved were retort stable physically as well as hermetically. OnTech utilized ANC's seam geometry and "chuck and role" profiles in order to reproduce the same seam quality, but with a multi-layer blow-molded container. This container manufacturing system provides a less expensive container but loses dimensional control (the flange diameter as the body must be continuous to a certain width). In order to achieve x, at the reduced cost of blow molding (vs. injection blow molding) a TRIZ technical contradiction was used:

Improving feature: Ease of manufacturing (32)
Degrading feature: Complexity of device (36)
Relevant inventive principles:  Inexpensive short-life of object (27), use of copies (26), segmentation (1)

Analyze Phase (With TRIZ)

Here, we apply the principles (generic solutions) to the technical contradiction. Segmentation stimulates the division of a mono-step manufacturing process (blow-molding with exact flange dimension, which is impossible) into a two-step manufacturing process (blow-mold and die stamp flange to correct dimension). This ideal solution is the answer to our flange problem: we get the needed dimensional accuracy utilizing an inexpensive process and a basic secondary manufacturing step.

Visually, the OnTech seam is as robust as the Omni-Bowl seam. The overlap, cover-hook and body-hook are comparable and in certain cases superior. (See Figures 4 and 5.)

Figure 5: Omni-Bowl Seam  overlap = .0387"

Seam quality was validated using microscopic dimensional analysis as well as immersion testing, micro-leak analysis and inoculated pack analysis. The importance of this preliminary piece of innovation becomes apparent when the container is sterilized via the retort process.

Retort Process

The retort process is a sterilization method by which the complete package is subjected to an elevated temperature (~250°F) for the amount of time required to stabilize the package temperature at the target temperature. The duration at this temperature is dependent on the required log reduction of bacteria in the foodstuff. A pressure vessel utilizing saturated steam is the means by which the heat is transferred to the package.

The pressure in the vessel is greater than 1 atmosphere so that T>212°F may be obtained. It is in this environment that the integrity of the double-seam is challenged. The fact that the OnTech container has three internal chambers exacerbates the problem of retort greatly. There are also four seams or sealing surfaces of the OnTech container subjected to the rigors of the retort process: two double-seams, a power-inertia weld, and a heat seal. The pressure differentials in the three chambers must be equalized in order to maintain physical structure.

Figure 6: Retort Temperature Profile

Figure 7: Temperatures of Components vs. Retort Temperature

As you can see from Figure X, the temperature inflections a', b', c', and d' occur at various times, t₁, t₂, t₃, and t₄, respectively. This compound inequality, t₄> t₃> t₂> t₁>t_retort; yields various physical contradictions during the retort thermal profile.

Figure 8: Macroscopic Zones of Conflict

As shown in Figure 8, philosopher Dr. Rudolf Carnap (1891-1970) described properties and the relation of physical things as follows: An object, or substance (S_x) as in Su-Field modeling, occupies a definite region of space at a definite instant of time, and temporal series of spatial region during the whole history of its existence. A substance occupies a region in the four-dimensional space-time continuum. A substance at a given instant of time is a cross-section of the whole space-time region occupied by the substance (a substance-moment). Carnap's insight into the ability to "slice" an event into time and space intervals is important for zone-of-conflict determination.

During the retort heating cycle, the puck water boils, the beverage boils, and the plastic body and cone soften. The headspace in the beverage compartment attempts to expand (the beverage is incompressible) and this exerts pressure on the end (A) (the location of a double-seam). The cone headspace expands exerting force on the cone wall (B). The puck headspace expands exerting force on the puck foil (C) (the water, again, is incompressible). A, B and C are positive pressures during the ramp-up cycle, where A>C>B (to the bacteriological death temperature). After the minimum hold time (at 250°F) the steam is collapsed and the pressure reduces to 1 atmosphere and the containers are cooled (water circulation in the retort vessel).

The cooling cycle, unchecked, creates vacuums, A^-1, B^-1 and C^-1, where A^-1> C^-1> B^-1. A^-1> wall strength of the cone, thereby causing the collapse of the cone (the cone volume changes from ~325cc to ~300cc post-retort). The cone collapse increases A^-1 and this vacuum A^-1> wall strength of the body, thereby causing body paneling. It is these two conflicts that require the retort cooling cycle to be controlled. (See Figure 9.)

Figure 9: Modified Retort Cooling Cycle

The controlled cooling cycle was the result of the following physical contradiction: You need the steam collapsed to end the thermal cycle, but you do not need the steam collapse because of the physical detriments. Therefore, utilizing the separation in time principle it is possible that the cooling process may be stepped (at precise t and P intervals) in order to allow the wall strength to recover before the vacuums, A^-1, A^-1, B^-1 and C^-1 are maximized.

The loss of cone volume post-retort occurred with pronounced variability and unpredictability. Rather than oppose this deformity (using principle 22 "convert harm into benefit") we designed to predict and control the deformity (see Figures 10-14). The control of the deformity allows us to minimize and control the vacuum, A^-1, so that A^-1 wall strength of the body wall. Controlling the cone deformity also allows us to insure enough unrestricted CaO expansion during hydration. The thermodynamic life-cycle of CaO is interesting in its elegance and beauty: it reverts to the form it was mined; therefore, no net loss to earth as an open system.

Figures 10 and 11: Cone Collapse Without  Structural Modification(s) to Control Deformity  or Assure Occurrence (Version 1)

Figure 12: Modification –  Control Deformity Under  Retort Loading (principle 22)

Figures 13 and 14: 93% and 96% of Retort Load – Deformation Is  Predictable and Controlled

Improve Phase (With TRIZ): Oxygen Ingress

The retort process forces retort water into the plastic of the body. This saturation of the polypropylene expedites the oxygen ingress problem (O₂ ingress to the beverage through the package that degrades the beverage flavor and allows bacteriological growth).

Thermal Conductivity

The low thermal conductivity of the cone (its primary constituent is polypropylene) causes various undesirable conditions. The design of the OnTech container includes a solid and liquid reactant which are co-mingled, when desired, in order to evolve energy (generated from a basic exothermic reaction) to heat the beverage. The container utilizes Fourier's Law of Heat Conduction (named for French mathematician and physicist Joseph Fourier (1768-1830)) to heat the beverage. This physical law states that two areas with different temperatures sharing a common non-insulated wall will reach equilibrium over time. At reaction initiation, heating begins and the temperature inside the cone becomes greater than the temperature of the beverage. In accordance with Fourier's Law the temperatures begin the equalization process by the transferal of heat through the cone to the beverage. The insulating properties of the cone (its primary constituent being a polypropylene homo-polymer) and the rate of heat absorption of the beverage, cause an inequality (transient heat transfer) in the dynamics of the heat transfer system (rate of energy created by the reaction is greater than the transfer rate through the plastic).

Figure 16: Thermodynamic Inequalities

The reaction is spontaneous and independent from the thermo-physical properties of the other elements of the system.

Figure 17: Experimental Case Demonstrating Potential Energy of  Reaction and the Realized Energy Transfer to the Beverage

For this particular experimental case the internal cone temperature reaches its maximum at 4 minutes and 40 seconds. The beverage, however, reaches its maximum temperature at ~8 minutes. The time difference (4:40 versus 8:00) is indicative of the thermal resistances between the energy source and the beverage. During the time interval at which the energy release rate of the cone core is maximized (~2:40-4:40) the rate at which the beverage is being heated is approaching its maximum. It is precisely these thermal resistances and the thermodynamic inequalities described that present the possibility of experiencing POMD (post operational maintenance drop) under certain conditions.

An effect of the thermodynamic inequalities is the thermal over-saturation of the polypropylene homo-polymer that comprises the cone wall. This thermal over-saturation, under certain conditions, could result in a cone breach.

The technical contradiction is as follows:
Improving feature: Strength (14)
Worsening feature: Temperature (17)
Inventive principles from contradiction matrix: Use of composite materials (40)

The application of principle 40 yielded the inclusion of ceramics and pitch-based carbon fibers to the standard polypropylene. The thermal conductivity of the composite polymers greatly increased the slope of heat generation and alleviated many of the failures associated with the previous configuration.

Figure 18: Polymer Comparison

The Finished Product

Figures 19-23: (l-r) Body and Cone; Metal End for Container's Bottom; Metal End for Container's Top;  Water Draining Onto the CaO; Energy Release From Cone to Beverage

Conclusion

The integration of The Theory of Inventive Problem Solving (TRIZ) with Six Sigma is absolutely necessary. The ability to solve problems in a closed-system and an open-system are vital to the optimum resolution of each problem. TRIZ and Six Sigma present the practitioner with the required capabilities to search for solutions in both domains.

About the Authors:

Michael S. Slocum, Ph.D., is the principal and chief executive officer of The Inventioneering Company. Contact Michael S. Slocum at michael (at) inventioneeringco.com or visit http://www.inventioneeringco.com.

Amir H.M. Kermani is an Industrial Engineer from Tehran Polytechnic. He has researched quality management techniques, particularly Six Sigma, DFFS, QFD and TRIZ. Kermani is among the few people who have tried to develop TRIZ knowledge and establish a TRIZ base in different industries (including chemical and petrochemical) in Iran. Contact Amir H.M. Kermani at Kermani (at) iiits.org.