Your test capacity management system should at least include these four essential attributes.
In my last several posts, I examined how managing chip test capacity has changed—and has not changed—over the past quarter century or so. These "chronicles" revealed, in part, that much has changed regarding the devices under test, ATE architectures, test methodologies, and test ecosystem structure —growth of the OSAT (outsourced semiconductor assembly and test) model.
What hasn't changed much, however, is the fundamental way in which test capacity is determined: a test specifier (chip designer or test engineer) tells one or more test providers (internal test groups or OSAT companies) the ATE model and configuration and test time that will comprise the required test capacity for a given device. With limited control over the specification of the required capacity, the test provider is therefore challenged to effectively aggregate multiple test requirements such that a high utilization and return on investment in the test equipment and related resources can be achieved. The resulting costs (and responsibility) of such test capacity management inefficiencies are shared by all participants in the ecosystem.
What you do to improve demand aggregation, utilization, and profitability in the test world? More specifically, what have the test stakeholders done to obtain such improvements? Interestingly, tools for managing test capacity haven't kept up with the level of accuracy and precision needed for the tight back-end margins of today’s mature semiconductor industry. While those in the test arena are no doubt applying significant resources and years of expertise to the test capacity management problem, there remains room for improvement. What would be the attributes of such a sophisticated test capacity management solution? Here are my thoughts on some of the essentials:
1. Accuracy and Precision. Legacy tools and techniques used to track test requirements and capacity typically view capacity as a set of “black boxes” and always fall short in capturing the many details necessary to accurately and precisely specify, match, value, plan, and trade (i.e. "manage") test capacity. Your capacity management solution should have the data model and import capabilities to capture and manage the unique and detailed taxonomies and attributes of all the major ATE models.
2. Immediately deployed, with no infrastructure changes. Given the advancements in secure cloud computing, you should be able to turn on your test capacity management system as simply as firing up your browser. No data servers to install, no applications to download, just log in and start seeing the benefits. Such cloud-based solutions can enable a number of other benefits as described below.
3. Accessible anywhere, on any device. Rarely are the key test capacity management decision makers gathered in one place and one time zone these days. All interested parties should all be able to securely view and manage their test capacity on their desktops at the office, on their phones at the airport, and on their tablets at home, collaborating both with internal colleagues and with external partners, and constantly communicating with their "industrial internet" of ATE.
4. Scalable with your business. The small test provider or specifier should not have to pay the same as a larger test provider or specifier for a test capacity management system. Ideally, the costs of your test capacity management system will be scalable on a per-system and/or per-device subscription basis, with no prohibitive one-time or annual enterprise fees, so that you can immediately see the benefits of the solution, and those benefits can scale with your business.
Of course, there can be many other attributes that might define your ideal test capacity management system, but these essentials should be included in your checklist as you look to improve your test capacity management efficiency and accuracy, and ultimately your profitability.