Survey provides profile of casting design software use.

This AMC survey shares benefits, costs and strategies so foundries may estimate the technology's potential impact on their operation.

Over the pest decade,the availability, capability and use of computer software for the design and analysis of castings has greatly increased. The use of CAD packages for part design and CNC programs for tooling and pattern production have been augmented by the development of design aids and molding simulation/analysis tools specifically targeted at the casting processes.

As part of an American Metalcasting Consortium (AMC) research project in software integration issues, the experiences of the early users of casting design software were studied to generalize the costs, benefits and nature of effective software use.

Casting Software Types

Casting software is broken down into four major divisions:

* CAD/CNC software--the familiar packages used to define and manufacture product and tooling geometries;

* special purpose casting software--computer programs that aid in determining gating and feeding system dimensions, sizing risers and estimating weights;

* solidification packages--programs that simulate the filling and solidification processes in the mold, and estimate defect and material property probabilities;

* other software--items such as production data management routines, which are often "home grown" on general purpose spreadsheets or database programs.

While the first and last types are important to the industry, the recent project investigated casting design integration, and thus focused on special purpose and solidification packages.

When the major special purpose software available in the U.S. and Canada was reviewed, it was discovered that all of these packages ran only on MS-DOS personal computers (PC). These programs were inexpensive, stand-alone systems that typically performed a specific calculation for a generalized casting design.

In contrast, solidification packages tended to be expensive, semi-integratable software that simulated or predicted several useful quantitative aspects for a particular casting design--and these programs were equally available on either the PC or the more powerful Unix-based workstation platforms. Due to this nature, the solidification package and computing platform combinations provided the basis for the best comparison of issues under study.

With the cooperation of 11 firms that supply the majority of the casting software used in the U.S. and Canada, a project team from The Univ. of Iowa surveyed 154 foundry users during the second and third quarters of 1995. The survey achieved a 48% response rate. These responses provide a basis for the following characterization of casting software use.

Benefits

The first task facing the project team was to try to understand the need for industrial casting software. The benefits actually experienced, and the degree of expectation fulfillment for the solidification software are the first aspects quantified.

Mean Benefits Realized from Software Use--Figures 1 and 2 attempt to capture the benefits available from use of casting solidification software. In Fig. 1, the mean percentage improvement and standard deviation range are shown for PC-based programs on seven performance measures. "Improvements" would mean decreases in required time or costs, and increases in yield. About 40-47% improvement was found in the part/tooling design and construction lead times. Average reductions of about 30% were realized in production time, labor costs and rework costs and an average of 25% improvement was found in the casting yield among the PC users.

[Figs. 1-2 ILLUSTRATION OMITTED]

Similar results were reported by workstation users (Fig. 2). Overall, the mean improvements were at least 25%, regardless of either performance criteria or computer platform.

User Rating of Software Performance--The survey questionnaire also asked the users to rate the performance achieved by their software on four individual criteria and on an overall basis (see Figures 3 and 4). Again, the results are broken down by platform and show similar trends. The assessment scale was anchored at 0 (very low) and 4 (very high); with 2 representing an average performance level. In general, the users reported reasonable performance in obtaining the expected results from their packages, with the workstation software being rated slightly higher than the PC programs for predictive accuracy and overall satisfaction.

[Figs. 3-4 ILLUSTRATION OMITTED]

Costs

Mean Implementation Cost--Figure 5 displays the mean implementation costs of solidification/simulation software by computing platform. The reported acquisition costs for the computing hardware form the base of each bar, followed by the software purchase, installation and training costs.

[Fig. 5 ILLUSTRATION OMITTED]

The major item of note is that the mean costs to effectively use the software are nearly six times greater for the workstation packages than for the PC programs. For the PC solutions, the major cost component is that of the software. Within the workstation environment, the equipment costs join the software costs to form the major cost hurdle. However, the cost difference is related to factors other than vendor platform pricing--the majority of the software employed on PCs in the study had fewer capabilities for predicting material properties, and didn't yet provide mold filling simulation. Such capabilities were being developed at the time of the study.

Mean Staffing Level--Another of the differences between the platform choices was found in the mean personnel requirements for operation. To adjust for the multi-user capability of the workstations, the mean man-years per seat were calculated for each of the platforms. Typical workstation staffing is nearly twice the PC level, at 1.06 man-years/seat (compared to 0.54 man-years/seat for the PC). This may be due not only to the differences in the information provided and expertise required of the workstation solidification user, but also to the differences in the employment strategies for the software.

Strategies for Software Use

As one would expect, differences in the strategies for effectively using casting software were found to be based on the nature of the casting business served. The following paragraphs describe the software preferences based on the business differences, and then the emergent software employment strategies.

Comparison of Users by Foundry Size--One of the most telling results was found when the employee size of the user firms was compared with the overall foundry industry figures on a percentage basis (Fig. 6). The first thing that is apparent (from the total height of the bars), is that the use of casting software is disproportionately lower among the smallest firms, and disproportionately larger among the medium and largest foundries.

[Fig. 6 ILLUSTRATION OMITTED]

A second look reveals that among the software users, the smaller firms tend to favor the PC based programs, while the largest foundries tend to prefer the workstation packages. A similar tendency was found when the firm size in gross sales was reviewed. The available personnel and capital within a firm appears to have a strong connection with the kind of software (platform) that can be obtained.

Comparison of Users and Industry by Sector Served--The frequencies with which respondents reported supplying a particular product sector were also compared against the frequencies found in the overall U.S. foundry industry. For the most part, casting software users tended to be found in proportion to the overall industrial distribution. Slightly lower frequencies of use, however, were found in the machinery and equipment sector, while slightly higher proportions were found in the "Other" sector.

Closer inspection of the returns showed that most of the those using solidification software cast products that required critical attention to the material properties of the casting, such as engine, hydraulic and aerospace components. The need for structurally sound or homogeneous product properties are prevalent among those suppliers.

Business Nature of Respondent Firms--The user firms were asked to characterize the business nature of their foundry operations. Almost 60% of the respondent firms reported that performing jobbing work for other companies was among their tasks. Around 23% reported their activities were targeted toward producing captive parts, and 18% were focused on providing design services. Several of the respondents reported that the analysis provided by the software was useful in improving the received cast part designs and in resolving casting design conflicts. These may be major reasons for the initial software acquisition.

Metals Cast and Casting Processes Employed--Workstation platforms tended to be employed more often in those firms that cast aluminum and nickel-based alloys (primarily investment cast, super-alloy foundries) than among those pouring cast irons and steels. PC based packages tended to be more prevalent among those that cast copper/zinc alloys. When viewed in light of the casting processes employed, about one-half of the 40% of sand casting respondents used PC-based programs. Of the other major processes, (about 23% diecasting, 18% permanent mold, 9% investment casting and 2% lost foam, with the remainder not reporting) about two-thirds or more of those users chose the workstation software.

Emergent Part Design-Software Strategies--The last of the primary aspects of software use considered here is the percentage of parts that were designed using casting software. When the responses are analyzed, about 45% of all software users designed 5% or less of their parts with the software; 25% of the firms designed 6-25% of their items with the software; 18% fell in the 26-50% of parts category; and few used the software on more than half of all their products. While virtually all of the PC users tended to use the programs in a small percentage of designs, the workstation users employed the software more often--evenly distributed in each of the first three categories--using it in up to 50% of their part designs.

In follow-up contacts, three user strategies for employing the software emerged:

1. To use the software to improve only those parts with high payback, such as difficult-to-cast or high production volume castings.

2. To use software occasionally to improve designs for new families of similar parts.

3. To use the software on most new designs, in order to reduce the risk of incurring the high cost of a die or permanent mold design failure.

Although examples could be found for each approach within a platform, most of the PC users seemed to employ the first strategy, while the majority of the workstation users tended to employ the latter two strategies.

Basis for Further Study

This report has attempted to encapsulate the costs, benefits and major implementation aspects of casting design software. Hopefully, by generalizing the information available from the main study and the summary presentation, a basis may be initiated for evaluating the casting software possibilities at the reader's foundry.

Previous articles ("A Deeper Look at Casting Solidification Software," L. Estrin, modern casting, July 1994 and "Study Evaluates Computer Modeling Programs for Die Casting," S.P. Midson, Die Casting Engineer, May 1994) covering the additional aspects of individual package prices and calculation speed provide a good (although somewhat dated) starting point for further investigation.

This area of foundry technology is extremely dynamic, and the benefits reported show a potential for profitable application. After all, 65% of the respondents questioned here indicated that they had used the software for less than two years.

This article was adapted from "Final Report: Metalcasting Lead Time and Cost Reduction Project" by C. Beckermann, G. W. Fischer, D.H. Jensen and S. Venkatasubramanian. To obtain a copy of the full report, contact the AFS Research Dept. at 847/824-0181.