A NIST research update on June 15, 2020 put estimated annual losses from preventable machinery-maintenance problems at US$119.1 billion for selected U.S. manufacturing industries. The study uses 2016 data, not pandemic-era results. For factory managers, its value is in identifying costs beyond the repair bill—not in promising what a particular monitoring system would save. [S1][S2]

The report separates US$57.3 billion in direct machinery-maintenance spending from the US$119.1 billion loss estimate. Within the latter, US$18.1 billion comes from unplanned downtime, US$0.8 billion from defects and US$100.2 billion from sales lost through delays and defects. These are different economic quantities. A missed sale is not the same as lost profit, and treating the entire loss estimate as cash available to fund new equipment would overstate what the study shows. Its scope is discrete manufacturing under North American Industry Classification System codes 321–339, excluding 324 and 325. [S2]

The practical choice is between different ways of deciding when to intervene. Reactive maintenance waits for a failure; preventive maintenance follows a schedule based on time or use; predictive maintenance draws on observations of equipment condition. NIST describes the last approach as requiring a manufacturer to decide what to monitor and which measurements should trigger action, then invest in hardware and software to capture and analyze the data. Buying sensors is only one part of that work. [S1][S2]

The survey offers a striking comparison, with an important limit. Establishments in the quarter most reliant on reactive maintenance reported that downtime consumed 13.0 percent of planned production time, against 4.0 percent in the least-reactive quarter. That is a comparison between establishments, not the measured effect of changing one factory’s maintenance policy. The authors warn that other measures can also reduce downtime and defects, and that many differences between the small groups are not statistically significant. [S2]

The national estimate is uncertain, too. NIST received 85 survey responses and excluded some because the respondents were not manufacturing establishments or had not completed key questions. The report gives a 90 percent confidence interval of US$43.9 billion to US$197.3 billion around the US$119.1 billion annual loss estimate. That broad range supports investigating the problem; it is a poor substitute for a plant’s own investment case. [S2]

An equipment proposal needs to account for the work around the machine. Scheduled maintenance can be excessive, using unnecessary labor and materials while keeping machinery idle. Predictive maintenance needs more than instrumentation: personnel may need training to interpret measurements and respond to specific warning signals. For a manager weighing these options, the practical implication is to budget for that work alongside the hardware. The budget needs to cover the people who interpret a warning and act on it, not just the device that produces it. [S2]

Source documents

  1. Research Suggests Significant Benefits to Investing in Advanced Machinery Maintenancehttps://www.nist.gov/news-events/news/2020/06/research-suggests-significant-benefits-investing-advanced-machinery
  2. Economics of Manufacturing Machinery Maintenance: A Survey and Analysis of U.S. Costs and Benefitshttps://nvlpubs.nist.gov/nistpubs/ams/NIST.AMS.100-34.pdf
  3. NIST publication record: Economics of Manufacturing Machinery Maintenancehttps://www.nist.gov/publications/economics-manufacturing-machinery-maintenance-survey-and-analysis-us-costs-and-benefits