When we first started selling products we discovered that they had a tendency to break and needed to be repaired or replaced. And we quickly learned three things:
- We can make money repairing stuff.
- Customers hate to pay to have stuff repaired.
- Customers really hate not being able to use the stuff while it is broken or being repaired.
These observations, and our experiences, let us to two critical conclusions:
- No matter how much maintenance you do it is impossible to eliminate every failure.
- The end-user cost of breakdown maintenance is normally more than any other type of maintenance.
These two conclusions resulted in the maintenance management evolution shown here where each service level provides increased customer value:
Breakdown Maintenance means, “When a product fails, fix it”. This is purely reactive and is like playing whack-a-mole; you fix something and then something else breaks and so it goes. And each time you have to fix the product, it cannot be used. In many B2B situations, the cost of lost revenue from not being able to use the equipment far exceeds the actual repair costs. Today, informed buyers do not want to operate in this mode – their jobs are at risk.
Preventative maintenance techniques generally are either routine or time-based. With this model a product is inspected, adjusted, and calibrated after a set amount of time or a fixed amount of use (routine). For example, your car manual may suggest a oil change every 5,000 miles or 6 months. The theory is that this service will prevent future, unscheduled failures. The problem with this approach is that it does not take into consideration the actual use history of each individual product. For example, a New York City taxi creates a lot more wear-and-tear in 5,000 miles than does a minivan driven around a typical middleclass suburb for the same distance.
Condition-based maintenance takes advantage of more modern instrumentation to indicate when an individual product is likely to need maintenance at a future time. For example, in all relatively new cars there is a “service engine soon” warning light that signals when the on-board computer detects either a true fault (low tire pressure) or a potential fault (low coolant level). The theory is that the warning will cause the operator to go to a qualified service center at a convenient time, where all the fault codes will be read and investigated until the “real fault” is identified. Depending the actual nature of the fault, the car may be repaired immediately or scheduled at a time convenient for the owner and the repair shop.
Predictive maintenance is condition-based maintenance on steroids. This method works with today’s highly complex, intelligent subsystems that are integrated into a product or network of products. Each intelligent component includes a computer that communicates to a central computer, either embedded in the product or located remotely. They communicate over the Internet using machine-to-machine protocols – this is the so-called Internet of Things (IoT). In many cases today, and certainly many more in the future, the central computer is not only monitoring subsystem status but also actual use. This information is analyzed using highly advanced algorithms that rely on the combined history of many similar products to project when failures will most likely occur. The system them schedule a maintenance appointment and make sure that a qualified technician, with the correct spare part(s) and tools will also be available. Working this way the maintenance work can get done before a failure actually occurs and at a time that minimizes the downtime of the process that uses the equipment. For example, if a piece of equipment in semiconductor fabrication line need a major repair the tool owner can also schedule other, deferred, repairs on other equipment. The line is down only once for required maintenance on a number of tools!
If your business is not headed in the direction of M2M and IoT, you will find that the leading-edge customers are not even considering you. You will be relegated to becoming a commodity.