Why Test Time Matters in Semiconductor Testing
Break down semiconductor testing revenue into chip shipments, test time, and the number of devices tested in parallel. Learn how test equipment sales differ from testing services and how to calculate equipment throughput.
Demand for semiconductor testing depends on the number of chips tested and the testing required per chip. Even with unchanged shipments, longer test times require more equipment hours, although parallel testing and program improvements can offset the increase.
Distinguish Test Equipment from Testing Services
Test equipment companies sell testing systems, while testing service providers are paid to test customers' products using equipment. Orders for one and utilization for the other are different indicators. Advantest explains both wafer testing and post-packaging testing. Identifying the stage in which a company operates helps distinguish demand to screen out defects during manufacturing from demand to verify finished-product quality. Contact components such as probe cards and sockets may also have replacement cycles different from the main equipment, so examine their revenue structures separately.
Are Longer Test Times Always Beneficial?
As chips become more complex and tests take longer, required equipment hours may increase. But if a service provider receives a fixed fee per chip, longer testing can reduce throughput and increase costs. You need to know whether a contract charges by equipment time or by product to determine the direction of the impact. For equipment suppliers, customers' increases in parallel testing and optimization of test procedures can also reduce additional purchases. News of improving chip performance alone is insufficient grounds for setting a testing revenue growth rate.
Separate Available Hours from Actual Billable Hours
When calculating testing equipment's hourly throughput, subtract time spent on product changeovers, calibration, maintenance, and retesting. If the test program changes, the processing speed of a previous product cannot simply be applied to the new one. To assess actual profitability, connect billable testing hours, rates, equipment depreciation, and labor costs. Higher equipment usage during new-product qualification may not yet have converted into billable volume-production testing. Utilization forecasts become more concrete when customer production schedules are assessed alongside the equipment configuration available to provide the service.
Worked example
Hypothetical example: Equipment that tests 4 devices at a time and takes 40 seconds per test cycle can theoretically process 360 devices per hour. If effective operating time is 80% of total time, throughput is 288 devices per hour. If test time increases to 50 seconds with other conditions unchanged, throughput falls 20% to 230.4 devices per hour. At an unchanged price per product, the revenue capacity of the testing service can also decline, so greater testing difficulty alone does not establish improved profitability.
What to check
- Distinguish equipment sales from testing services
- Check test time per chip and the number tested in parallel
- Distinguish pricing per product from pricing per hour
- Account for time lost to changeovers, maintenance, and retesting
Sources and primary materials
This is educational material about research methods. For an actual company decision, verify the filings and primary sources for the relevant period.