Concept Development

 

 Quality Function Deployment Method (QFD)

 Function Analysis

 Morphological Chart

 Concept Generation / Concept Selection

 Final Concept Selected

QFD

     The quality function deployment method is defined as the strategic arrangement throughout all aspects of a product of appropriate characteristics according to customer demands. The QFD is essentially concerned with the translation of customer requirements into engineering characteristics. The QFD method involves identification of the customer requirements in terms of product attributes and then determining the level of importance of each one. Drawing a matrix of product attributes against engineering characteristics and rating the relationship between them will complete this area of the process. Engineering characteristics may have relationships between each other, this is called the house of quality. The table below illustrates the QFD for the yeast cultivator.

Quality Function Deployment

Back to top

Function Analysis

     It is important to break up the design problem into several functions in order to gain a better understanding of the problem. Function analysis was used to decompose the cultivator problem into three main functions: environment, sampling, and structure. These were then broken down further into many individual sub-functions. A function tree was then used to display these functions and sub-functions because it was a clear way to identify the relationship/interdependency between each sub-function. This function tree is displayed in Figure 1.

Function Tree

Back to top

Morphological Chart         

     Developing a design requires that the design problem be increased into several different concepts before the final design is narrowed to one solution. A method for accomplishing this task is by forming a morphological chart. A morphological chart is a list of means to achieve the function desired. As many methods as possible for each function should be entered. Once finished, the morphological chart contains the complete range of all theoretically possible different concept forms for the product. This complete range of concepts consists of the combinations made up by selecting one function at a time. The morphological chart for the yeast cultivator is illustrated in Appendix A. From the morphological chart different conceptual designs were generated.

Morphological Chart

Back to top

Concept Generation / Concept Selection

Concept Generation

     Generating conceptual designs is achieved by selecting a mean for each function and combining them to form concepts. These combinations are existing solutions; some new solutions, and some impossible solutions. During the process it is necessary to identify potential solutions for further consideration. Using this method three concepts for the cultivator were developed. A schematic drawing of each concept was then created to visualize each individual concept. These drawings are below.



Concept #1 Box Shaped Cultivator




Concept #2 Cylindrical Yeast Cultivator with external magnetic field




Concept #3 Similar to Concept 2 except sampling method is different



Concept selection (Pugh Method)

      Our concept selection utilized a decision matrix method (Pugh method) which is based on comparing the needs or wishes with the concepts. Concepts chosen for this matrix must meet two requirements. First the concept must be feasible to both the customer and engineering group. Second the concept must satisfy all of the customer's needs. After the concepts has passed the two screening requirements they are evaluated with respect to each other using a fixed reference (datum). The Pugh method tests each concept for completeness and helps invent new concepts from previous ones. A criterion must be chosen for comparing concepts. The demands are chosen for the first comparison criteria. If all the concepts rank the same for the demands then the wishes are used as the criteria. The wishes use a weight factor from 0-10 as a way to control the importance of the wishes. If the demands are not all at the same importance level then they can be placed into the matrix as well. In order to rank the concepts they must be rated against a benchmark design. This design is either a previous design or a concept that is an obvious solution to the problem. The concepts are rated to be either better than (+), equal to (0), or less than (-) the datum concept.After the concepts are scored the scores are tabulated. The pluses and minuses are counted for each concept. Then the difference between pluses and minuses are counted; this is the overall total. Finally the sum of each score is multiplied by the weight of the criteria; this is the weighted total. The overall total is used to see where the individual concept is lacking or where its strong points are with respect to the criteria. A criterion that is ranked with the same score for most concepts should be looked at closer in order to generate better concepts. In this case the criteria may not be under stood to its fullest extent. To grasp a better understanding of the concepts this process should be repeated with the highest-ranking concept as the new datum. Using this method, the concept chosen for the yeast cultivator was concept number #2. After carefully inspecting the negative aspects of this concept using the pugh chart modifications to the design were required. The final pugh chart for the yeast cultivator can be seen below.



Pugh Chart

Back to top

Final Concept Selected

     Through further analysis of the concept, several negative aspects were found to be inappropriate for the design. Positive aspects from other concepts were incorporated in this concept to produce the best possible design. Another reason for modification to this concept was due to revisions made in the need statement and in the specifications. There were several drawbacks in the selected design in which altercations were needed. The Photon Multiplier (PM Tube) found in the bottom of the system was discovered to be very sensitive to magnetic fields and was subsequently removed from the bore of the magnet. By removing the PM Tube the shutter was eliminated. This required that fiber optic cable be used to relay the luminescence signal from the yeast to PM Tube. In order to magnify the luminescence, a lens was attached to both ends of the fiber optics. Another feature that was modified was in the sampling tube. A constraint on the amount yeast being tested and the vertical magnetic field variation led to the redesign of the tube. The new tube design consists of two concentric tubes in which one is allowed to slide within the other. A small suction on the inner tube allows a regulated sample to be taken; then the inner tube can be extracted from the device in order to externally test the sample for contamination. Insulation was added due to a large temperature gradient between the medium and surface of the magnet bore. The insulation was installed on the outside and inside surfaces of the water jacket as well as on the top and bottom of the concept. Due to the added insulation the controlled water bath to the medium was eliminated. A detailed drawing of this concept is below.


Final Concept



Back to top

Last Revised: 2000-12-20