Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing
Reexamination Certificate
1998-10-15
2002-04-30
Grant, William (Department: 2121)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Product assembly or manufacturing
C700S080000, C700S110000, C714S025000, C714S048000
Reexamination Certificate
active
06381508
ABSTRACT:
The present invention relates to the technical field of industrial production, and more particularly of the productivity of “machine” lines.
In industry, machines are controlled by a “programmable logic controller” (PLC), which controls certain functions of the machine. The PLC receives orders from the operator via, in the simplest configurations, a control panel.
In the prior art, “man-machine” interfaces have been devised whose function is to translate complex operations and functions of the PLC, and hence of the machine, into simple functions and manipulations. Thus, the complex functions become accessible to the operator and/or the latter executes or controls them more quickly.
Among the known interfaces are those which include the display of machine breakdowns and machine states, as well as those offering, for example, a debugging aid, which can include a diagnostic help function.
Also known are audit lists or “checklists,” computer-aided maintenance of the machine, which is an external function that makes it possible to issue intervention orders for any part suspected through analysis of being liable to become defective within a given time period (so-called preventive maintenance), and the so-called conditional maintenance that occurs when a critical threshold is reached (vibrations, oil quality, frequency of breakdowns, etc.), which triggers an intervention.
It is possible, according to a first aspect of the invention, to classify breakdowns according to different methods.
For example, it is possible to differentiate
the non-critical alarm messages (prevention of the breakdown),
the critical alarms that result in a machine shutdown, with attribution of the down time to the breakdown involved,
the so-called reset breakdowns which, because of their seriousness and/or because of the moment in which they occur, require the resetting of one full operation on a production line (for example, a restart of the entire sterilization operation),
and the breakdowns known as “going down,” which make it necessary to restart the entire production, that is, to restart all of the operations preliminary to a production start, with a potential loss of production in the case of a fragile product, a breakdown of a cold chain, etc.
According to this first aspect, it is therefore possible to classify breakdowns according to various systems, identify them, and construct algorithms for handling every type or group of breakdowns.
“Breakdown” is intended herein to mean any machine incident of any importance, whether capable or incapable (examples of “undocumented” breakdowns), as desired by the designers, of generating an “alarm.”
“Alarm” is understood to mean any signal generated by a “breakdown” for the purpose its being handled.
“Machine” or “machine line” or “production,” etc., is intended herein to mean any type of industrial or semi-industrial production comprising a certain degree of automation and an alarm management system, or capable of being so equipped, in any type of industry.
There is also a known system for optimizing productivity known as “TPM” or “Total Production Maintenance,” a complete maintenance system at the production level, which generally distinguishes between two types of breakdowns, chronic breakdowns (those that occur with substantial frequency), and unexpected breakdowns (which are rare, but costly in terms of machine down time). Although very useful, and more complex than described above, this system, like others, is limited to a statistical handling of breakdowns and to the initiation of certain preventive or reparative actions.
These known systems can be more or less sophisticated, but they have a common point, which is that they do not assist in the precise AND automatic calculation of productivity.
In fact, in their algorithms, these systems identify a breakdown, classify it into one of the categories that have been imposed on them, and attributes the machine down time (or any other indication of production disruption) to this breakdown.
From this attribution follow statistical calculations and recommendations for actions such as the replacement of a certain part, diagnoses of normal or abnormal wear, etc.
It is therefore a very systematic approach that is limited to determining actual states such as a breakdown, and to performing calculations resulting from these actual states.
According to the invention, what is proposed is a production control system that integrates into the breakdown handling module a step for recognizing the initial or primary breakdown of a downstream or secondary breakdown having resulted in an alarm.
According to the invention, the machine down time is attributed to the primary breakdown. Thus, the derived calculations direct actions to the primary or “real” causes of the breakdowns, and no longer to their visible, but “secondary” effects, which would not have been produced without the occurrence of the primary breakdown.
Hence, a certain “culpability” criterion is assigned to each breakdown, possibly with weighting coefficients that are within the scope of one skilled in the art, on each type of machine line.
As an example, let us consider a rolling bearing whose lubrication is obviously controlled, for example by means of a sensor of the flow of lubrication fluid. Let us accept the hypothesis that this sensor stops functioning; therefore a “pressure sensor breakdown” is issued. A temperature alarm threshold will also be crossed if the security system has been well designed, issuing a second, “temperature sensor breakdown.”
If, for the restart, it is necessary to operate a panel, etc., that is also being monitored (access panel, electric switch, etc.), third or fourth alarms are registered by the system. If the overall reaction is too slow, a mechanical failure can occur, for example the deformation of an overheated mechanical element: fifth alarm.
In the standard systems, each alarm is simply treated as such, and the down time is attributed to it. Thus, the maintenance operator is informed of two, three, four or five breakdowns caused by only one breakdown, the first. The statistical calculations are correct, but their interpretation is wrong; thus, in the above example, the statistics could suggest that the design of the mechanical element should be reviewed, whereas only the pressure sensor should have been implicated.
This results in erroneous diagnoses and predictions of wear, and therefore in totally unnecessary or largely premature interventions, while the main cause is not eliminated, or at least not quickly enough.
This results in a decrease in productivity which could have been avoided. The invention, by assigning a “primary responsibility” criterion to a breakdown, makes it possible to pinpoint the breakdowns that must really be handled, and to eliminate from the statistics the breakdowns that did not occur spontaneously. Thus, the manager can order the handling of the main causes of breakdowns, and no longer order useless interventions.
Thus, a productivity gain threshold is crossed, since the invention does not involve a more or less marginal improvement in the sophistication of an algorithm, and hence a more or less substantial refinement of its calculations, but operates according to a radically different concept, which is to handle only the breakdowns whose handing is useful.
Full automation of the monitoring of an operation is also obtained.
The practical embodiment of the invention, being based on the above concept, therefore requires an algorithm that recognizes the links between the various elements “under control.” Herein, an element under control designates any part, any sensor, etc., for which the designer of the machine, in accordance with his plan for security and production control, wishes its failure to trigger an alarm.
In the above example, the algorithm must therefore know that if the fluid pressure is abnormal, the temperature sensor will display an abnormal value (with a reaction time that is itself calculable by means of an algorithm) which will trigger an alarm; that it is necessary to replace the senso
Arent Fox Kintner & Plotkin & Kahn, PLLC
Grant William
Rodriguez Paul
LandOfFree
Process and equipment for optimizing production does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process and equipment for optimizing production, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process and equipment for optimizing production will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2825718