“Draft:Disruptive Power of Repair Management Method DPRMM”的意思、由来-开放百科全书

Many names have been given to numerous models used to evaluate and estimate machine downtime and when to replace capex that costs more to maintain than to operate. These include PMO, which stands for Preventative maintenance, and planned maintenance optimization. Another subset includes PdM, which is Predictive Maintenance.

Where PdM concentrates on models to determine when equipment needs to be maintained, PMO and capital expenditure (CAPEX) replacement models used to evaluate the economical quality of equipment (capex) lifetime cost to a system. OTRME and DPRM are separate subjects, DPRM is a DpR system, and concentrates on preventing downtime due to repair, OTRME concentrating on when to replace the equipment (capex), and uses the data from DPRM (DpR) to make a decision.

Introduction

PMO/PdM presents a predicted versus actual cost a equipment (Capex) has to a system. The predictive cost is what is considered to be the lifetime cost broken down into hours based on the expected lifetime span and is predicted at the time of procurement as a purchasing and investment decision. The actual cost of the capex is based on real time performance of the capex during its lifetime cycle after it has been installed. PMO and PdM employ various models, with various names, to view, record and project the time to replace capex based on the frequency of malfunctions and the growing costs of repair as well as the detrimental effect the repair downtime has on the system.

History

As a subset of Reliability Engineering, both PMO and PdM have their history entrenched in the emergence of reliability engineering. Practical reliability can be traced as far back as 1920's, where research into reliability and quality were on a parallel path.^[1] During WWII, the use of reliability became a standard for product reliability.^[2] . The report characterized the field of reliability as being the product of improving component reliability, establishing qaulity and reliability requirements on suppliers and collecting field dat that will enable finding root causes for failure.

After WWII slight changes were made along the way to how the military managed their reliability testing, including standard 781 and the handbook 217, which was the work of RCA and became known as Mil Std.217. During the 80's SAE published their SAE870050 reliability document for the automotive industry. The 90's saw he introduction of reliability in their standard ISO9000. From the 90's onwards reliability has become a standard of operation for every industry and with the constant evolution of computing power, software solutions are proving practical applications in every industrial sector.

Determining When to Replace Equipment (OTRME)

A whole section of systems engineering is dedicated to optimizing the replacement of capex. This subject area is where quality, operations and finance meet in one algorithm, all dedicated to finding the cheapest and most lucrative solution to replacing the old capex. One such model is the optimum time to replace malfunctioning equipment (OTRME) model takes the current acceptable models (RCM, TPM), differentiates between repair time and maintenance, and adds two new repair-based factors that are not present in the decision-making process. These are the disruptive power of repair (DPR) and the malfunction ripple effect (MRE) that a repair downtime has on a whole system.

CAPEX Life Cycle

Every capex, irrespective of its complexity, has a life cycle. The life cycle is based on the effectivity of providing a quality service to reach its purpose.

When choosing a new piece of equipment, the procurement process uses many variables to determine viability. The variables include the product functions that are set by the primary stakeholder. The primary stakeholder is the individual or department that sets the attributes of the equipment, including the equipment’s dimensions, functionality, and productivity based on the various physical attributes and production process that the equipment must integrate into. These attributes are the stakeholder defined ranges or “constants” that cannot be changed without the stakeholder’s consent.

The procurement variables that the primary stakeholder allows the buyer to consider when creating a list of makes and models to choose from are price, supply-chain, installation, training, maintenance, warrantees, and in some cases, substitutes that might be considered where they prove to perform better than the original concept. Procurement professionals have a wide range of tools to use for determining the equipment they intend to purchase. Porters five forces model^[3] is one of the most comprehensive, and linear performance price (LPP)^[4] indicators are another method used for determining whether less complex equipment and raw materials should be purchased based on their stakeholder demands.

One of the important variables that help in making the final decision for purchasing equipment is the estimated product purchasing price and lifespan cost. These values are imported into a projected final product pricing, that will give the company the estimated cost and projected profitability of the equipment. This estimation is at the core of the OTRME. The estimated cost versus the actual cost is based on a projected lifespan estimated cost versus the actual yearly costs. OTRME calls this the baseline cost per hour. It is significant since projected and estimated profitability are based on this figure.

In most models, the purchase price of a new piece of equipment is considered a major factor in the decision-making process. In fact, one extensive paper discusses this issue in great detail (Kärri, 2007). OTRME discounts this concept, and considers the process of procuring a new replacement for old malfunctioning machinery a separate decision-making process that should not have any impact on the rationale to replace capex. This is best observed when new technologies replace old ones. The cost of introducing a new technology would warrant such an important place in the decision-making process^[5]. OTRME only evaluates the impact a malfunctioning machine has on a process and decides when it is best to replace it. In the event of replacing old equipment with new due to the impact the frequency of repair has on the production process, the cost of the replacement machinery has no impact on the decision-making process for when to replace, it is only a factor for deciding what is the replacement make and model, and how much it costs to procure.

Maintenance is a standard procedure to keep equipment operational, as defined by Merriam Webster (2018b)“the upkeep of property or equipment” and repair “to restore by replacing a part or putting together what is torn or broken” (Merriam Webster, 2018c) is when a part of the equipment is damaged or must be replaced. Equipment undergoing maintenance is scheduled into a plant operations model, and preventative maintenance is a daily, sometimes pre-and post-operational procedure to maintain the equipment and general maintenance is performed according to a schedule when the equipment is “dissected” for a scheduled maintenance check^[6].

Repair is when damaged equipment demands unscheduled stops and is determined by as an irregularity. Many companies incorporate risk management for instigating contingency plans when equipment malfunction, these risk contingency costs are not included in any published equipment replacement model to date. Therefore, all current models do not include the use of replacements or substitutes that take over operations when a piece of equipment is in repair. This extra cost shortens the actual economic lifespan considerably.

Malfunction Ripple Effect (MRE)

The MRE is value based on the overall costs a repair process creates in a system. When a machine is in repair downtime, either the machine is an isolated repair incident and the only effected process is the one being stopped, or, the product that must be processed might require the transfer of process to another machine. In this instance, the repair downtime will cause a ripple effecting other process for different products. This ripple must consider:

The total value received is the overall cost to the system caused by the downtime.

The Frequency of Downtime (FD)

The FD is the variable that comes into force when calculating the full cost of repairs. Since maintenance is a scheduled operation, in most cases companies do not schedule replacement equipment when factoring standard and preventative maintenance. The downtime is an accepted norm of operation. With repairs, the downtime is not normal, and as such, most companies will immediately factor into their process a risk contingency plan. These risk plans include the use of a replacement to maintain the flow of work. The FD is critical since it raises the cost per hour significantly. For example, when a machine malfunctions once, and the repair takes a week, and the machine operates flawlessly for the rest of the year, the cost might be high, but the machine is not subject to replacement. If, however, the machine malfunctions ever few days and the downtime is above four hours each time, while the cost might be less than a month’s downtime, the disruptive effect, and the collateral costs, point to the machine being problematic and demands a replacement. Therefore, higher FD is more critical than the length of downtime, since it is an indicator that the equipment is failing to perform on an increasing basis.

The Disruptive Power of Repair (DPR)

The DPR has been downplayed and even overlooked in all models to date. Although it is a much-discussed issue in maintenance engineering research (Life Cycle Engineering, n.d.). When analyzing a production process; as a piece of machinery malfunctions, its “job” load must be distributed to another piece of equipment. If there is an identical machine, then the situation can be monitored for job balancing. If, however, there is no identical replacement, then the job must either be delayed or transferred to an alternative machine that can perform similar functions if that is possible. When viewing the first option, the less complex the machine, the easier it is to balance in, unless the job loads are such that balancing is not an option. With more complex machinery, for instance, a CNC machine (Akturk and Avci, 1996) that requires a jig or apparatus to hold the material for operation and software (code) that defines the automatic milling operation, then downtime to set up machinery must be factored into the overheads, and this can and usually does significantly increase the cost as well as damage the job balance. We also deemed it important to add the fact that repaired equipment performs less than pre-repaired , which reduces the quality of performance every time the machine is repaired (Chen et al., 2015). This is called the disruptive power of repair and its estimation is complex due to the number of additional dependent variables that come into play when trying to transfer production from one line to another. DPR is given an important value, depending on the nature and complexity of the machine that is needed. The frequency of downtime with more complex systems is more important than the cost since each downtime has a ripple effect on the whole system as presented by the MRE.

The OTRME Model

The OTRME model comes into play once the device has been fully installed and is ready for use. The device begins its life cycle at this point, and its economic value is based on the income or savings it generates from its use. Productivity equipment, diagnostic devices, and operational tools are all income valued. This means that the tool is used in a procedure that is estimated and given an income value. The device used in the procedure is given a percentage of cost variance based on the projected usage of the device across all the procedures it might be used for.

OTRME is effective in all productivity tools and equipment since they can be directly related to an income source.

The Model is based on over 8 subsets of data that are arranged in a modeling system, but are adapted as equations for use in standard software such as Excel and Matlab. The final outcome of the model is:

OTRME allocates initial cost minus salvage value to each month, together with averaged monthly repair cost in the future, to get the monthly ownership cost.

Estimating, Predicting and Managing Equipment Reliability (DPRMM)

DPRMM is a method devised to standardize an efficient approach to how a system will manage a repair process.

The DPR management method (DPRMM) was devised as an additional system to complement the OTRME, in which both models estimate and forecast when the capex must be replaced as well as devise a method to optimize efficiency in how systems manage repair downtime effects, effectively lengthening the life cycle time of equipment by reducing the costs of capex downtime for repair.

DPR is a much-discussed issue in maintenance engineering research (Life Cycle Engineering, n.d.). When analyzing a production process, when a piece of machinery malfunctions, its “job” load must be distributed to another piece of equipment. If there is an identical machine, then the situation can be monitored for job balancing. If, however, there is no identical replacement, then the job must either be delayed or transferred to an alternative machine that can perform similar functions if that is possible. When viewing the first option, the less complex the machine, the easier it is to balance in, unless the job loads are such that balancing is not an option. With more complex machinery, for instance, a CNC machine^[9] that requires a jig or apparatus to hold the material for operation and software (code) that defines the automatic milling operation, then downtime to set up machinery must be factored into the overheads, and this can and usually does significantly increase the cost as well as damage the job balance. It is important to add that repaired equipment performs less than pre-repaired, which reduces the quality of performance every time the machine is repaired^[10]. This is what is termed the disruptive power of repair and its estimation is complex due to the number of additional dependent variables that come into play when trying to transfer production from one line to another. DPR is given an important value, depending on the nature and complexity of the machine that is needed.

The first splash or the epicenter of the downtime is the malfunctioning equipment or machinery. This is allocated as the center of the ripple. The first wave that comes out of the center is the production units immediately affected by the downtime of the equipment. The second ripple is the number of cost centers that get involved in discussing the issue and finding a solution; the third ripple is the effect the repair process and the malfunction have on the management of the company. The final ripple is how the malfunction affects the inventory and supply chain process, and whether the downtime is going to impact the customer.

The downtime for repair is not just how long it takes to repair a malfunctioning piece of equipment or machine; it is a whole matrix of interlinked causes and effect that contribute to the cost of repair, taking the cost from other important actions and transferring them to the repair process. This means that a system will not operate at 100% efficiency when downtime is present.

Naturally, the more important the machine or equipment within a process, such as a linear or vertical process that a malfunction will stop, the greater the impact of the repair on the system. This impact is called the disruptive power of repair. DPR takes the number of times a machine breaks down as the frequency of downtime and the effect that each downtime has on a system as the downtime ripple effect. These three factors contribute to estimating the optimum time for replacing a malfunctioning piece of machinery, that could be could be causing the company much more damage than they realize.

There are a number of ways to estimate the value per hour of a company, the value of a cost center and the value of a specific tasks manpower costs. However, the simple version is to ask a simple question: If your company just stood for a day, what would it cost?

This question can mislead since there are three kinds of companies, service, retail, and production. A service company only needs to factor in all its expenses, and in most cases does not have much capex. A retail company has a lot of inventory that is usually dynamic, so the correct way to estimate the inventory value would be to use the average cost of inventory method. A production company is much more complex to estimate since it has many hidden corners of inventory, such as maintenance, as well as a lot of capex and WiP. With these differences come either an easy estimation or a hard one, so no matter which type of company you have, you need to estimate the overall cost your business will spend when it is standing still for one day. This is what is called the daily cost of operation, and it details how much a company costs every hour.

The next variable that is used to estimate are the cost center costs; these are defined sectors within a company, such as sales, maintenance, administration, production, engineering, etc. Since we already have the overall cost of a company’s hour, all that is needed is to categorize costs centers and then allocate the resources to each category from within the companies defined cost per hour.

The final variable is to ascertain is the cost of the heads of cost centers, these are only hourly manpower rates, but necessary to estimate the time used per cost center head to deal with a situation. A cost center head is a person in charge of a specific task, and this requires that specific tasks be defined and allocate resources. So, the maintenance of an air filter would require specific resources; this would be defined as a cost center head for the maintenance of an air filter, which is the cost of the person dealing with the problem, without the resources used to solve the problem.

The variables are then assimilated into a system framework that generates the total cost of repair within the organization. In most cases, such a framework would be a project management software solution. The integration of DPRMM is simple and that is why it is so popular and is used in thousands of companies around the world.

Top Companies that integrate PMO (OTRME)

IBM is first in rank due to the size of its workforce and its global popularity. IBM has a PMO solution called Predictive Maintenance and Quality (PMQ) which is hthe product of its “cognitive intelligence engine” IBM Watson. The IBM PMQ monitors, analyzes, and reports OTRME data, and the output is presented as a health-score. Some of IBM's leading PMQ success are Kone’s elevators or DC Water’s Hydrants.

SAP is a German software that incorporates PMO in its software packages. SAP OTRME/DPRMM is called “Predictive Maintenance and Service” and is integrated onto many systems including Kaeser Kompressoren and Siemens. SAP Leonardo IoT Portfolio comes with the OTRME/DPRMM solution built in.

Unlike IBM and SAP, Siemens is a device manufacturer, and with over 100 years of experience, this giant's devices sits in just about every machine and building around the world. As such, the PMO solution is built into the Siemens DNA and PMO data is collected through its devices on a routine basis. Some of its leading customers integrate OTRME?DPRMM solutions through Siemens projects, including NASA Armstrong Flight Center (cooling systems).

Another software giant, Microsoft introduced Azure which is a popular public cloud platform used by industrial IoT solutions and PMO's. PMO is becoming more and more a remote predictive solution, where OTRME/DPRMM solutions sit on clouds and record from afar.

GE has two PMO solution providers: GE Measurements which sits in the condition monitoring hardware field, and GE Digital which provides software and analytics for PMO. GE developed the Predix platform, which is the foundation to Asset Performance Management (APM), GE's name for OTRME/DPRMM. GE has integrated APM with BPs oil and gas production operations

References

1. ^{{cite book|last1=Juran, Joseph and Gryna, Frank,|title=Quality Control Handbook, Fourth Edition|date=1988|publisher=McGraw-Hill|location=New York|page=24.3}}
2. ^{{cite book|title=Reliability of military electronic equipment;report|date=1957|publisher=United States Department of Defence|location=Washington}}
3. ^{{cite journal|last1=Porter|first1=M.E.|title=The Five Competitive Forces that Shape Strategy|journal=Harvard Business Review|date=2008|volume=86|issue=1|page=78-93}}
4. ^{{cite journal|last1=Newman, W. R. & Krehbiel, T. C.|title=Linear performance pricing: A collaborative tool for focused supply cost reduction|journal=Journal of Purchasing and Supply Management|date=2007|volume=13|issue=2|page=152-165}}
5. ^{{cite journal|last1=Nair, S. K. & Hopp, W. J.|title=A model for equipment replacement due to technological obsolescence|journal=European Journal of Operational Research|date=1992|volume=63|issue=2|page=207-221}}
6. ^{{cite journal|last1=Colledani, M. & Tolio, T.|title=Performance evaluation of transfer lines with general repair times and multiple failure models|journal=Annals of Operations Research|date=2011|volume=182|issue=1|page=31-65}}
7. ^ Mathematical_optimization#Optimization_algorithms
8. ^Mathematical_optimization#Optimization_algorithms
9. ^{{cite journal|last1=Akturk, M. S. & Avci, S.|title=An integrated process planning approach for CNC machine tools|journal=The International Journal of Advanced Manufacturing Technology|date=1996|volume=12|issue=3|pages=221-229}}
10. ^{{cite journal|last1=Chen, X.; Xu, B.; Yang, Z.; Chen, F. & Meng, G.|title=Reliability Model for subsystems of CNC Machine Tools based on the Repair Degree|journal=6th International Conference on Manufacturing Science and Engineering|date=2015|pages=28-29}}
11. ^{{cite web|title=The Top 20 Companies Enabling Predictive Maintenance|url=https://iot-analytics.com/top-20-companies-enabling-predictive-maintenance/|publisher=IOT Analytics}}
12. ^https://www.ibm.com/developerworks/library/ba-pmq-devicewise/index.html
13. ^https://help.sap.com/saphelp_globext607_10/helpdata/en/66/1586e3547611d182cc0000e829fbfe/frameset.htm
14. ^https://www.energy.siemens.com/co/en/services/automation-controls-electricals/preventive-maintenance/
15. ^https://docs.microsoft.com/en-us/azure/iot-suite/iot-suite-predictive-walkthrough
16. ^https://www.ge.com/digital/blog/prevent-evolve-profit-future-field-services