|
|
 |
INTEGRATING PLASTIC PROCESS SIMULATION INTO THE CORPORATE STRUCTURE: MARKETING AND MANAGEMENT IMPLICATIONS
Geoffrey Engelstein, GR Technical Services, Mountainside, NJ
Introduction
Computer-Aided Engineering analyses are powerful tools for diagnostics and troubleshooting of the injection molding process. However many organizations do not take advantage of these technologies because they do not fit neatly into the corporate structure. Often consultants are used on an ad hoc crisis basis, where costs are higher and potential solutions limited. Integration strategies are discussed for the part design firm, toolmaker, and molder. Implications for financial, engineering, marketing, and management structures are examined, along with cost, timing, and responsibility issues.
Marketing and Management
Management is often leery of CAE for a variety of reasons. The benefits of the technology may not be clear. There is a desire not to 'rock the boat' if the engineering department is performing adequately. The organizational changes required may not be understood. But CAE implementation must come from the top, both for financial and management reasons. There will also be a resistance on the part of engineering to integrate new tools into the normal development cycle, and a strong commitment from management is essential to make sure the system is used to its potential and doesn't become an expensive paper weight.
However, many marketing advantages accrue to companies that make the commitment and implement plastic analysis. The first is improved time to market. Analysis reduces average development time and makes it easier to stick to schedules by avoiding major problems.
In addition, the ability to predict aesthetic defects in advance, like weld lines and blush, allows those typically outside the development cycle, such as sales and marketing, to be directly involved in design decisions. Having everyone 'buy off' on a project early in the cycle obviously leads to fewer conflicts near the launch date. This advantage is often overlooked given its intangible nature, but it yields concrete benefits.
Also important, especially for tool shops and custom molders, is the 'high tech' factor that simulation presents to the potential customer. Being able to present cutting edge technology as an integrated part of your services has definite marketing value and serves as a differentiator against the competition.
Financial Considerations
Before tackling integration questions, it is useful to examine the financial impact of an analysis system. Unfortunately a direct cost/benefit analysis is problematic at best. While the costs of a system are fairly straightforward, savings associated with simulation are more difficult to measure, as they arise from problem avoidance. A complete justification of the costs of simulation is outside the scope of this paper, however a quick examination will prove useful.
The first thing that needs to be considered is the direct cost of the hardware and software. Depending on the analysis sophistication required, hardware costs will range from $10,000 to $30,000. Similarly, software costs range from $10,000 to $30,000 depending on capability. Obviously it is much less expensive to set up a system if the hardware is already in use. The computer power required is equivalent to that for a CAD system. In addition, maintenance costs of 5%-10% per year are typical for computer service and software upgrades.
The cost for the analyst depends on whether he is performing full time simulation, or is just adding this onto other duties. Regardless, a certain number of simulations need to be performed in order to justify the expense of training and maintaining the system, as well as keep the analyst reasonably efficient. In our experience, a minimum of one analysis per month is necessary to stay current on the software. Any less than this and the analyst will have to relearn the software each time a new simulation is begun, resulting in longer lead times and general frustration.
Consultants will typically charge between $1000 and $5000 per analysis. Many factors affect internal analysis cost, but with sufficient simulation volume, in-house simulations will be approximately half that of hiring a consultant.
The financial benefits of simulation lie not in reduced tooling times, but in reduced debug times. Typically four weeks are scheduled after tooling completion to sample and make minor modifications to the mold. However the actual debug time can vary from zero (the rare mold that works perfectly right off the bat) to a very long time, when molds have major problems and large reworks ranging from part design changes to gate and waterline relocation. We have found that simulation cust the average debug time in half, while reducing the number (and hence cost) of post-tooling changes. These savings however, will be experienced over many molds. On some, the initial tool design turns out to be the best, so the simulation added only peace of mind. On others, though, the 'killer problem' that would have led to costly mold rework will be avoided. So drawing financial conclusions from only a few analyses is misguided.
Another area of cost savings that is difficult to quantify is the optimization of parts. Many times parts that are less than perfect (excess gate blush, for example) are accepted just to hit a launch date or stay under budget. Simulation can fix these problems within the original schedule. The overall improvement in part quality caused by these small, pre-tooling modifications cannot be understated.
Types of Simulations
Plastic process simulations fall into three loose categories: Filling analysis simulates the injection, packing, and holding stages of the cycle. It is useful for determining flow pattern, weld locations, blushing, injection pressure, potential short shots, and other injection-related parameters. It assumes adequate mold cooling to maintain a constant mold temperature.
Cooling analysis adds details about the mold base to the simulation. Waterlines, baffles, bubblers, inserts, parting lines, and other effects can be added to the model. The term 'cooling analysis' is somewhat of a misnomer, as the injection phase is also simulated. However unlike a pure filling analysis, the effects of mold temperature variations are incorporated into the plastic flow. Hence it is sometimes call a 'coupled fill-cool analysis'. In addition to the results from the filling analysis, core and cavity temperatures, coolant temperature rise, and other cooling system related parameters can be examined.
The last type of simulation is the warp analysis, which determines the deformation of the part after ejection. The simulation evaluates density variations from the filling phase and temperature variations from the cooling analysis to determine internal part stresses.
Simulation and the Development Cycle
Because of the varied types of simulations that can be performed, it is often unclear how analysis fits into the typical development cycle. In order to examine this question, we first look at a very basic development.
The three broad steps that occur during the development of an injection molded part are:
- Part Design
- Tooling
- Molding
These steps are performed by the design engineer, tooling engineer, and manufacturing engineer respectively. This is an admittedly simplistic but still useful characterization. The three stages can all be performed within a single organization or spread out among many different companies, like custom molders and tool shops.
Simulation is typically used at two points within this cycle. After product development filling analysis is used to develop gate locations, runner sizing, and to evaluate the potential for aesthetic or processing problems (such as blush, weld line location, or injection pressure). Any design changes are then made to correct any problems uncovered.
The more complex cooling and warp analyses are usually not performed at this early stage, as cooling line layout has not been established. It is difficult to create water lines in a vacuum, as mold elements like ejectors and slides will have a major impact on their potential size and location.
Once the overall runner/gate layout is established, tool design commences. Once the framework of the mold has been established on paper, potential cooling channel designs are simulated with a coupled fill-cool. Also, more detailed process data is developed, such as coolant temperatures and cycle times. If there is a problem the mold and part designs are modified to correct it.
This process is shown in Figure 2. The initial filling analysis is more of a product validation, while the later fill-cool analysis is a mold and process simulation. The main problem, as indicated by the question marks on the figure, is who performs these analyses.
Who is the analyst?
The process analyst requires many skills. The most important of these is knowledge of plastics. The computer tools, while very sophisticated, require a knowledgeable human to interpret the data and take corrective action. It is critical to understand what physically is happening to the plastic during injection, be able to anticipate problems and focus on key areas, and understand which changes will improve the parts or make them worse. The computer will not make material recommendations, or suggest that a certain wall thickness be changed. It will merely point you in the right direction.
Tooling knowledge is also important, so the analyst can evaluate which part and mold changes will be possible and what the cost impact will be. Detailed knowledge of precisely how to build a mold is not required, but without basic understanding of how a mold is put together, and what the design limitations are, much time can be wasted simulating changes that will be impossible to implement.
Obviously in order to make effective use of the simulations the analyst must also be familiar with computers, and the operation of the software. While much of the software available today is 'user-friendly', icon driven, and well-documented, if the analyst is comfortable with the system it will make the simulations that much easier to perform in an efficient manner. However, it is usually easier to teach someone computer skills than convey the plastics and tooling knowledge.
So the ideal analyst will understand plastics, design, tooling, molding, and computers! People like this are difficult to find, so a team-oriented approach is indicated. An analyst that is supported not only with high quality software but also with experts in other areas of plastic molding will be able to perform more efficiently and more accurately. The ability to harness this knowledge with the organization is the single most important determinant of whether simulation will be successfully integrated or not.
The Integrated Company
A company that performs, or at least coordinates, all steps of the development cycle has the best opportunity to bring these diverse strands of knowledge together. Simulation is often thought of in conjunction with concurrent engineering. In one sense simulation forces an organization like this to adopt a concurrent approach as the analyst serves as the focal point for design, tooling, and manufacturing personnel and knowledge, uniting the upstream and downstream development phases.
Most often the analysis functions are merged in with the design group, since that is where the CAD systems are. If the designers are on CAD they have the computer knowledge to easily operate the software. In addition, they intrinsically understand what design changes would be acceptable, minimizing exploration of blind alleys. The problem with this setup is that designers tend to be insulated and unresponsive to feedback from manufacturing. If this approach is taken it is imperative that cross-functional teams be stressed and actively used. Otherwise many of the main benefits of analysis (cycle time improvement, for example) will not be realized.
Depending on the size of the organization, a better solution is the dedicated analyst. Since the software tends to be expensive, rather than providing each designer with simulation capability, all analysis is channeled through a few people. In addition, the efficiency of the analysts is improved as they become more familiar with the tricks and idiosyncracies of the software. By focusing the analysis on a small group it forces them to reach out to the design and manufacturing groups, fostering the desired integration.
The drawback to the dedicated analyst is the bottleneck that can be created if many parts are brought to market simultaneously. This can lead to release delays or bypassing of simulations altogether. Also, in smaller organizations the dedicated analyst will not be cost effective, both from a personnel standpoint (the additional salary), as well as the additional computer hardware required.
Design Firms
Dedicated design firms include not only independent design groups but also large corporations which design products in-house but go outside for tooling and manufacturing. These groups have many of the same problems discussed above when integrating simulation. Very often sufficient knowledge of mold construction and plastic processing is lacking, leading to under- or mis-utilization. If analysis is going to be performed on the design level, it is critical that the designers solicit suggestions and share results with those who will build the tool and mold the parts.
On the plus side, design firms often already have a substantial investment in CAD computers, making the additional investment as low as possible. In addition, most simulation software today is integrated with the design packages so that generation of a model suitable for analysis is a simple task.
Mold Makers
Outside of integrated firms, toolshops are in the best position to incorporate simulation into their organization. Because the two phases of simulation bracket mold construction, it makes sense for the toolmaker to perform the simulation. They have knowledge of both molds and processing. The main deficiency is in understanding the critical parts of the product, but they are used to working with customers to learn this information and create molds that will produce parts that meet their aesthetic and functional needs.
Another potential hurdle for toolmakers is the lack of high end computers that are typical in design firms. However, the penetration of computer-aided machining (CAM) into the toolshop is increasing the level of computerization. In addition, tooling engineers are now familiar with computer surfacing and other techniques which are common to both simulation and CAM. Some extra time will be involved if the model has to be developed from scratch, but transferring data between dissimilar CAD systems is becoming easier.
Customer Molders
Of all of the organizations discussed, custom molders will have the toughest time justifying and integrating simulation. Other than simulation there is no reason for them to have high end computers in place, so implementing the hardware will be more expensive, along with the added training and cost required for new personnel. Some molders are implementing CIM systems, but these typically do not have sufficient power to run simulation software.
Also, the customer molder is furthest down the development chain, so it can be doubly difficult to be involved in early design decisions, which is critical for successful implementation. If the analysis is only performed on pre-existing tooling which is handed to the molder, the potential remedies which are available are drastically slashed. However there are many other positive forces driving molders to become involved in upstream decisions, such as the growing popularity and necessity of concurrent engineering, that make early simulations less of an adjustment.
Consultants
The final implementation consideration is the use of consultants. Consultants, if knowledgeable, can be a valuable addition to a design team if in house simulation is unavailable. However it is important to make sure that they are knowledgeable about molding, and understand the challenges of your product. Also, because they are operating outside of the organization, additional efforts must be made to ensure that they are part of the development team, and involved in the communication loop.
Responsibility
While plastic process simulation has come a long way since its introduction twenty years ago, it is still a simulation. That means that the results will occasionally deviate from what was predicted. Material data, poor modeling, and other problems can lead to inaccurate results. Although sophisticated users of analysis tools experience problems in only about 1% of the cases, it is important to be prepared. When molds don't work properly, much finger-pointing can result, although certain accepted standards of responsibility prevail. However, simulations can change the rules.
For the integrated company responsibility for the mold is a simple issue. Only one entity is involved. However, for stand alone entities like design firms and tool shops the issues are less clear. For example, let's say your firm designs products, but hires other companies to build the molds and manufacture the parts. Normally, the toolshop would be responsible for designing and building a mold that produces acceptable parts. However, if you perform the analysis yourself and hand the results to the toolmaker showing where the gates need to be and where the waterlines should run, that responsibility has now shifted onto your shoulders if the mold does not perform as anticipated. Likewise, a toolmaker may demonstrate via simulation that a mold will perform with a given cycle time. Most of the time it will perform as advertised, but if not, the responsibility shifts back to the toolmaker.
On the flip side, many toolshops today are being confronted with customers who want guarantees of certain cycle times. In these cases it is imperative that simulation be used to give a level of comfort to all parties.
Simulation is also very useful for assigning appropriate responsibility after a problem is discovered, cutting down on finger-pointing. While definitely not the most effective way to use simulation, one of its chief uses is as a troubleshooting tool. Once problems are experienced, analysis is performed to pinpoint the cause. It can be determined if the problem is tooling, part design, or processing. In its troubleshooting mode, however, the solutions available to the analyst to solve the problem are limited.
Conclusion
Simulation is an extremely powerful tool. However, to be successfully integrated into the organization a multi-disciplinary approach needs to be taken, involving design, tooling, and production. A company that handles all phases of plastic part development and production is best equipped for the technology, but other organizations, especially mold makers can also benefit.
|
|
|
