In the Manufacturing Sector an hour of plant downtime can cost up to 30.000 (in some cases even more).
This regards the lack of production: however, it is good to remember that an unplanned DOWNTIME can cost up to 10 times compared to a scheduled stop (for maintenance, control, testing, etc…).
The reason is very simple, parameters that "derive" from plant shutdown itself must also be taken into consideration.
Let's briefly summarize all the costs
- Equipment costs (calculated annually with constant unit price)
- Cost of labor: Takes into account the total cost of direct and indirect labor with benefits, and includes a share of all general positions in plant and support managers.
- Product cost: The cost per unit of production at each stage of the process together with the units per hour at the machine center / profit can indicate the value of product lost in an accident.
- Startup cost (per machine, line, cell and profit center) – Includes energy costs, set up (materials and labor), percentage of production reduced (units lost), scrap produced (including recycling costs and/or scrap value), quality ( inspection and rework costs), as well as other start-up costs.
- Bottleneck cost: the impact on downstream equipment at each stage of the process.
- Sales expectations : do not respect Downtime costs (per event)
- Cost per hour: Calculate and record the time from the first occurrence of the equipment failure to the time the equipment returned to full production.
- Reduced production
- Waste
- Service costs: The costs of temporary fixes to be made until the permanent fix is re-established.
- OEM, advice and assistance: Annual fee or the estimated cost per year for support during downtime.
- Tooling: Calculate the replacement or rework cost for tools (per occurrence).
- Replacements and repairs: shipping costs, repairs, tests, etc….
99% of UPTIME can also be read as 3,5 days of annual machine downtime: do you want to calculate what it would mean for your plant?
ServiTecno's proposal to increase AVAILABILITY is…
Increasing the availability of systems up to the achievement of High Availability (99,99%) or Fault Tolerance (99,999%) is a question of method: redundant resources starting from the field is essential and today (thanks to the spread of virtualization) on the "center" it is possible to obtain exceptional results with technically simple and economically affordable solutions.
Apply redundancy solutions with virtual machines it is simpler and certainly cheaper than basing your own AVAILABILITY' on the power of the servers: in this case (as for some solutions based on software platforms) it is not so much the switch between a machine (physical or not) that delays the restarting of the system, but rather the restart of the application.
With the solution EverRun by STRATUS TECHNOLOGIES distributed in Italy by ServiTecno, these latencies come even completely eliminated, since in reality there is no distinction between the primary virtual machine and the secondary one: the application runs in fact on two distinct environments but aligned to the millisecond thanks to a dual dedicated network card.
The two "servers" can also physically reside in two distant locations (provided there is adequate connection infrastructure) using the SPIT SITE option.
But EverRun doesn't stop at this: while running on two virtual machines, the operator user will see the application as if it were only one. It will also have the possibility to check the health of the physical components of the servers and their connections with the field through the STRATUS console: when one of these is transferred (for example the hard disk) the system continues to work using crossed resources and at the same time report the problem so that it can be resolved as soon as possible.
Adding value to the solution is the fact that EverRun does not limit itself to "putting" a single application under Fault Tolerance, but can bring together all the virtual machines that I intend to protect within the same application, just as if it were a large container of virtual machines: to balance everything it is possible to create the distinction between machines in High Availability and machines in Fault Tolerance, depending on the criticality of the application itself.