The company was under contractual obligation to deliver a quick release top nozzle (QRTN) to several customers. This is a complex subassembly costing approximately $1,700 each for a major piece of capital equipment. The current product, a removable top nozzle (RTN), is needed in case a product has to be repaired because of a failure. While this repair is a low probability occurrence, the cost of downtime is very high. If the product can’t be repaired within a reasonable period, there are significant operational costs involved.
The current RTN allowed repair within a reasonable time if all things went according to plan. However, there was considerable time and cost involved in mobilizing equipment to support the repair, in setting up the equipment to support the repair, removing and replacing the top nozzle, and tearing down and demobilizing the repair equipment. There were additional problems as follows:
In order to meet contractual requirements as well as develop a more competitive design for other potential customers, a project was initiated to develop a QRTN. A previous QRTN project was cancelled late in the Detailed Design Phase after the estimated cost increased to a level significantly above the RTN cost (+80%). Additional contractual commitments caused the company to restart the project after several years. A multi-functional core team was created, and a decision was made to use the newly-developed NPD process, target costing, value analysis, and quality function deployment (QFD) to help the team accomplish its objectives. In preparation for this program, hands-on QFD, target costing and value analysis training was provided to team members, several key functional managers, and other potential users of these tools. In addition, a QFD consultant was engaged to help the team apply these tools.
QFD logically begins with understanding customer needs and using these needs to drive the development process. These customer needs are referred to as “voice of the customer” (VOC). Instead of solely relying on the marketing organization to define these needs, team members met with customers to gain a first-hand understanding of their needs. It was recognized that there were multiple customer voices to consider with this project:
Prior to meeting with customers, the team obtained a briefing on the prior QRTN effort to familiarize them with some of the issues. Meetings were scheduled with engineering personnel at the customers that had a contractual requirement for the QRTN. The initial meetings were intended to probe and primarily listen for the customer needs. A list of questions to ask and a meeting agenda were prepared to guide these interviews. The teams were briefed on asking “Why” to understand the fundamental customer needs. Several team members were involved in each of these interviews, and they typically talked with multiple people in each customer organization. Interview notes were summarized and distributed to all team members. In addition, these meetings were discussed in detail in subsequent team meetings to assure a complete understanding of each customer’s perspectives. Contractual requirements were reviewed. A teleconference with the marketing representatives for the customers was held to gain their perspective.
The customer visits also provided good insight into the design of two of the competitor’s products and an in-house design from one of the customers. This proved to be a good opportunity to gather competitive assessment information and provided the information to develop rapid prototyping models and/or sketches of competitive joints.
To gain a better understanding of the voice of Services Division, the team watched a video of the repair process to gain a first-hand understanding of the issues in top nozzle removal and installation. Data was also assembled (Table 1) detailing the time requirements for Services Division to perform its job with the RTN and estimates or goals with the QRTN.
|Current RTN Design
(3 people required)
|QRTN Design Estimates
(2 people required)
|Top Nozzle Removal||2||0.5|
|Top Nozzle Installation||3||0.5|
|Post Repair Equipment Tear Down||15||2|
The team met to compile all of the requirements. Brief natural language expressions of customer needs (e.g., “maintain locked joint”) were written on “post-its” and placed on the wall. Some questions were asked to clarify meanings and to determine “Why”. The team used Affinity Diagramming to organize related statements into logical clusters. Besides organizing the customer needs, this also facilitated the following:
Initially, there were thirty-three statements of customer needs. By organizing these requirements with affinity diagramming, examining each of these requirements, and asking “why” to understand the fundamental need, this list was consolidated into twelve customer needs. Table 2 shows examples of this consolidation process.
|Original Customer Needs||Consolidated Customer Need||Comment|
|Thimble rotation resistance
Precludes skeleton damage
No loose parts during joint operation
|Joint reliability during repair||By asking “Why”, consolidated customer needs back to the basic need.|
Low cost tooling
Minimize tool mobilization
Minimal removal force
|Simple tooling||Consolidated requirements because simple tooling generally would lead to lower cost and cost was not deemed as important a customer need. Couldn’t directly control mobilization with the design, but concluded that simple tooling would achieve this same objective. Minimal removal force also related back to simple tooling.|
|Fast nozzle removal/install
Repair on critical path
No loose parts during joint operation
Minimum number of joints
Maintain alignment during repair
|Fast repair||By asking “Why”, consolidated customer needs back to the basic need. Many of these lower-level needs then became technical characteristics.|
At first, the team did not understand the value of the data dictionary. However, as numerous questions came up over the course of this project, the data dictionary proved invaluable. See the final data dictionary.
Once these customer needs were identified, they were organized into a product planning matrix. Based on the initial customer meetings, priorities were assigned to each requirement using a 1 to 5 scale. The information gained on competitive products as well as prior internal competitive assessments was used to look at competing joint designs from a customer perspective and develop an initial Competitive Evaluation. Actual hardware, rapid prototyping models, and sketches were used to support this evaluation. The Competitive Evaluation was done for six joint designs: the current RTN design, the previous QRTN design, a joint company/customer design, a customer design and two competitor’s designs. This information was reviewed with customers. They were satisfied with the stated customer needs in the matrix. They provided feedback to make minor changes to priorities and the competitive assessment.
The customer needs, priorities and competitive assessment were reviewed and a product plan was developed. The next step was to develop technical characteristics of the products to respond to customer needs. This was one of the most difficult steps for team members and required a good deal of facilitation and thought. The criteria for the technical characteristics were:
As these technical characteristics were developed, their definitions were also defined in a data dictionary to assure a common understanding among team members. Again, this was referred to many times during discussions. Examples of the technical characteristics defined in order to satisfy customer needs are shown in Table 3.
|Customer Need||Technical Characteristics||Discussion|
|Visually verify locked joint||Joint verification time||The removal and installation process is done under conditions where the joint can’t be directly seen, but is observed via video. It is required to visually verify the joint is locked upon reinstallation. This is videotaped. The nozzle develops a black coating in operation making visual verification difficult. It was observed that the contrast of new parts and having part features that protrude above the surface help with verification. However, stating these as technical characteristics was not global. Therefore joint verification time was determined to be the most appropriate technical characteristic.|
|Fast repair||No loose parts during operation
Operator lifting force
Joint lock verification time
Maximum removal installation time
Protrusion above adapter plate
OD collet to insert clearance
No moveable tooling parts
|A variety of factors support the customer need for fast repair. Loose parts contribute to additional handling time and potential time if part is dropped. If lifting force is greater than what a person can lift, additional equipment and set-up time are required. The two time-related criteria are straight-forward goals. If protrusions exist above adapter plate, additional time is required to lift tool over protrusion to properly seat the tool. The greater the OD collet to insert clearance, the easier it will be to access and pull parts of the product that require replacement. Moveable tooling parts may contribute to additional operation time.|
|Joint reliability during operation||Minimum joint activation force/torque
Use of approved material
In-operation loose parts simulation/ analysis
|The customer need relates to the QRTN joint remaining locked during operation and a concern that small, fine-featured parts may corrode and be broken during operation. The first related technical characteristics establishes a minimum force required to unlock joint to avoid the joint coming unlocked due to dynamic forces during operation. The next two characteristics relate to use of materials and simulation to assure that parts do not corrode or break during operation.|
As each technical characteristic was defined, the goal for that characteristic was also established . maximize, minimize, or equal to the target. Next, the relationship to each customer need was established. The relationship defines the degree (strong – weight of 5, medium – weight of 3, or weak – weight of 1) that the technical characteristic satisfies the customer need. It does not define the potential interaction of the characteristic with the customer need. The team often stumbled in this regard; they wanted to define negative interactions between the customer need and the technical characteristic. For example, a number of technical characteristics were seen as having a negative impact on “Minimum price increase” causing the team to want to establish a relationship. Frequent reminders were made that this was not the purpose of this relationship. After the relationships were established, the importance rating of each technical characteristic was calculated.
As technical characteristics were developed, preliminary target values were defined to assure that there was a measurable characteristic. In a few cases, the difficulty in defining a meaningful target value caused the team to redefine the technical characteristic.
The next major step was to define interactions among the technical characteristics. To keep the interactions manageable, the team ignored the minor interactions and even the moderate positive interactions. The strong negative interactions (“N”) were of particular importance to manage. Examples of the strong negative interactions are shown in Table 4.
|Technical Characteristic||Opposing Technical Characteristic||Comment|
|Target cost||Guide thimble mis-location from true center||To minimize guide thimble mis-location from true center requires tighter tolerances that drive up cost. This potential conflict can addressed with tolerance analysis and careful allocation and management of tolerances.|
|Inactive joints||Guide thimble stresses
Adapter plate stresses
Joint component stresses
|As the number of active joints is reduced, the stresses on all of these components increase. After discussing this issue and considering the potential number of times that an assembly would be reconstituted and joints overridden, it was decided to reduce the target value for inactive joints from 4 to 2.|
|Protrusion above adapter plate||Joint lock verification time||While the protrusion above the adapter plate needs to be minimized to prevent interference with other components in the customer’s system and prevent impacting cooling flow, some protruding features aid in verifying a locked joint and reducing the verification time if the protruding features can be kept below the height of this target value. This potential conflict can be managed through team awareness of the issue.|
The product planning matrix was finished by establishing target values once the interactions and importance of each technical characteristic was understood and establishing an estimate of technical difficulty for each of these characteristics at the target value. See the final QRTN product planning matrix
The product planning matrix was used to identify areas for follow-up and investigation. For approximately half of the technical characteristics, required follow-up steps or areas of investigation were identified to finalize target values or assess achievement of target values as concept alternatives were explored. Examples are shown in Table 5.
|Technical Characteristic||Target Value||Planned Action|
|Minimum joint activation force / torque||32 in-oz / X lbs||Establish minimum force / torque requirements|
|Deflection limit under 4g load||0.025 ins.||Investigate basis of 0.025 in. specification|
|OD collet-to-insert clearance||Zero on nominal stack-up||Review and optimize OD collet tool for max guide thimble clearances|
|Inactive joints||2 joints||Determine history of failed joints from the Service Division’s history|
Once the technical characteristics of the QRTN were understood, product design began. The first step was to develop concept alternatives. In order to meet target costs and the defined technical characteristics, creative solutions were needed. This process began with brainstorming. A brainstorming session was scheduled with not only the QRTN team members, but approximately 6 other people that had experience with nozzles and past QRTN efforts. The brainstorming session occurred over a six-hour period. It began with briefing on the product planning matrix and the technical characteristics and design issues that needed to be addressed. A function tree diagram was prepared to focus people’s attention on the functions that needed to be prepared rather than starting from the point of existing design concepts.
The group was then divided into teams of three to four people to brainstorm and then develop concept alternatives for the QRTN joint. Each team then presented their concepts to the rest of the group and discussion ensued. The QRTN team then took those concept alternatives and further reviewed and discussed them. The concept alternatives were organized into six families and the most promising concept in each of these families was further discussed and developed by a group of sub-team members.
The team used a concept selection matrix with the technical characteristics as decision criteria to screen these concept alternatives and down-select to four alternatives. An objective of the down-select was to yield two traditional alternatives and two more radical alternatives to carry forward. One of the problems recognized was that the criteria were not “leveled” and, therefore, insufficient weighting was given to target cost. Adjustments to the weightings were needed to address this. Sub-teams then further developed these four alternatives.
The concept selection matrix in conjunction with engineering analysis and development of product cost estimates will be used to select two final design alternatives which will be carried forward to the Preliminary Design Phase. One of the first steps in this phase will be to prepare a deployment matrix to further define critical characteristics of key components.
In parallel with the effort to gather the voice of the customer, analysis was done to develop a target cost. Marketing was asked how much would customers be willing to spend to obtain a QRTN feature vs. the RTN feature. Marketing agreed that the QRTN feature would be of interest to customers and that they should be willing to pay more for it. However, it became apparent that this feature would more likely be offered to maintain competitiveness at no additional cost or would be folded into overall pricing so that it was not apparent what the price difference would be.
The company provides the product under warranty. When a product requires repair, the company pays the Service Division to perform the repair. Any time savings in performing the repair because of the QRTN would be reflected in lower warranty costs. Data had been gathered in Table 1 showing the planned time savings that a QRTN design would offer over an RTN design. The Service Division representative on the core team provided data on the crew size and labor costs. Data was also collected on the number of repairs performed over a several year period. This was used to determine frequency of occurrence. Based on all of this data, the expected value of warranty savings with a QTRN design was calculated. This was added to the RTN baseline cost to yield the target cost established for this project. This calculation is shown below.
Target Cost = RTN Baseline Cost + Warranty Cost Savings
Warranty Cost Savings = Frequency of Repair Occurrence x QRTN Cost Savings
QRTN Cost Savings = Labor Rate x (RTN Manhours – QRTN Manhours) [See Table 1]
This target cost was included in the Product Planning Matrix as a target value.
Once the work was begun on concept development, the GA SEER DFM software was used to support cost estimating. The model was validated by developing a model of the existing RTN and comparing the model cost estimates with the current RTN costs. A model of the previous QRTN design was also established to validate the model against previous cost estimates. The team was unfamiliar with the cost model and did not appreciate how it could be used to support the process. Meetings were held to discuss this and the model expert was assigned to support the team. He met with the team to define a preliminary parts list and a preliminary process definition (manufacturing process steps or routings) for each alternative. The intent was to develop preliminary cost estimates to support:
The project was designated as high priority for the Division and established as a “heavy-weight” program with a key functional manager designated as the project manager. The initial staffing consisted of:
Because of the challenges with achieving target cost, two additional manufacturing engineers were assigned to support the team on a part-time basis. In addition, since this was a pilot project, two additional members of the value engineering team participated in the meetings to observe the process and support deploying the process to future programs. One of those individuals was the lead designer on the past QRTN program and became heavily involved with this program, providing briefings and contributing to the discussions and analysis. It soon became apparent that additional designers were going to be required to support the program and this individual along with another part-time designer were formally added to the project team. The other member of the value engineering team was also drafted. It is important to recognize the resource intense commitment that is required to support a wide investigation of alternatives and active manufacturing involvement in the early phase of a program.
This entire process took twenty weeks from the start of training to selection of the final concept alternatives for development. The most time consuming activities were collecting the voice of the customer in the beginning and the preliminary engineering development of the concepts including the cost estimates.
The product is entering the final development and testing phases. Estimated costs at this point are approximately 15% above the RTN cost and 5% above the cost target. However, this is significantly less than the previous QRTN development where costs were estimated at 80% over the RTN costs. As a result, this project is considered a major success and a successful demonstration of QFD and target costing.