Data processing: artificial intelligence – Knowledge processing system – Creation or modification
Reexamination Certificate
1999-07-15
2003-01-14
Starks, Jr., Wilbert L. (Department: 2121)
Data processing: artificial intelligence
Knowledge processing system
Creation or modification
C706S060000, C702S182000, C702S185000
Reexamination Certificate
active
06507832
ABSTRACT:
TECHNICAL FIELD
The present invention relates generally to printing systems and more particularly to a method and device that identifies conditions leading to, and that decreases the occurrence of, web breaks within a printing system.
BACKGROUND ART
Large-scale printing systems, such as rotogravure printing presses, feed a continuous web of material, typically paper, through printing machinery that forces the web into contact with one or more rotogravure printing cylinders which, in turn, print images onto the web in a standard manner. Thereafter, the web is cut into individual pages or signatures which are collated to produce newspapers, books, magazines, etc. A common and recurring problem in large-scale printing systems is the occurrence of web breaks, which happen when the web tears while the web is being fed through the individual components of the printing system. Upon the occurrence of a web break, the printing system must be shut down, the torn web must be dislodged from the individual components of the printing system and then the web must be re-fed through the printing system, all of which takes a considerable amount of time and results in wasted paper and ink. Furthermore, in some instances, web breaks may result in damage to components of the printing system.
While web breaks are a common problem in the printing industry, the reasons or conditions that lead to the occurrence of any particular web break vary widely. In fact, web breaks may be caused by different factors or by different combinations of factors at different times in the same printing system. Generally, web breaks are avoided by having a local expert, such as a printing press foreman, oversee the press conditions and make suggestions for changes based mainly on past experiences with web breaks, trial and error and general rules of thumb. While some of these approaches are successful in decreasing the incidence of web breaks in the short term, web breaks usually reappear later with very little indication as to the real cause of the reappearance. Furthermore, while local printing experts are usually capable of determining the general cause of any particular web break after the web break has occurred and, moreover, are generally capable of altering press conditions to eliminate a particular cause of a web break in the short term, there is no guarantee that the altered conditions will not result in further web breaks for other reasons or that the press conditions suggested by the local expert will be implemented in the press for a long period of time.
It has been suggested to use an expert system to determine the causes of problems, such as web breaks, within a printing system. In particular, the above-identified parent application, which issued as U.S. Pat. No. 5,694,524 on Dec. 2, 1997, on which this application relies for priority, is directed to the use of a decision-tree induction analysis that identifies conditions leading to a particular result, such as web breaks, within a printing system. In general, expert systems are used to mimic the tasks of an expert within a particular field of knowledge or domain, or to generate a set of rules applicable within the domain. In these applications, expert systems must operate on objects associated with the domain, which may be physical entities, processes or even abstract ideas. Objects are defined by a set of attributes or features, the values of which uniquely characterize the object. Object attributes may be discrete or continuous.
Typically, each object within a domain also belongs to or is associated with one of a number of mutually exclusive classes having particular importance within the context of the domain. Expert systems that classify objects from the values of the attributes for those objects must either develop or be provided with a set of classification rules that guide the system in the classification task. Some expert systems use classification rules that are directly ascertained from a domain expert. These systems require a “knowledge engineer” to interact directly with a domain expert in an attempt to extract rules used by the expert in the performance of his or her classification task.
Unfortunately, this technique usually requires a lengthy interview process that can span many man-hours of the expert's time. Furthermore, experts are not generally good at articulating classification rules, that is, expressing knowledge at the right level of abstraction and degree of precision, organizing knowledge and ensuring the consistency and completeness of the expressed knowledge. As a result, the rules that are identified may be incomplete while important rules may be overlooked. Still further, this technique assumes that an expert actually exists in the particular field of interest. Even if an expert does exist, the expert is usually one of a few and is, therefore, in high demand. As a result, the expert's time and, consequently, the rule extraction process can be quite expensive.
It is known to use artificial intelligence within expert systems for the purpose of generating classification rules applicable to a domain. For example, an article by Bruce W. Porter et al.,
Concept Learning and Heuristic Classification in Weak
-
Theory Domains
, 45 Artificial Intelligence 229-263 (1990), describes an exemplar-based expert system for use in medical diagnosis that removes the knowledge engineer from the rule extraction process and, in effect, interviews the expert directly to determine relevant classification rules.
In this system, training examples (data sets that include values for each of a plurality of attributes generally relevant to medical diagnosis) are presented to the system for classification within one of a predetermined number of classes. The system compares a training example with one or more exemplars stored for each of the classes and uses a set of classification rules developed by the system to determine the class to which the training example most likely belongs. A domain expert, such as a doctor, either verifies the classification choice or instructs the system that the chosen classification is incorrect. In the latter case, the expert identifies the correct classification choice and the relevant attributes, or values thereof, that distinguish the training example from the class initially chosen by the system. The system builds the classification rules from this information, or, if no rules can be identified, stores the misclassified training example as an exemplar of the correct class. This process is repeated for training examples until the system is capable of correctly classifying a predetermined percentage of new examples using the stored exemplars and the developed classification rules.
A patent to Karis (U.S. Pat. No. 5,521,844) discloses a case-based expert system that may be used to aid in the identification of the cause of a particular problem, such as a web break, in a printing system. The expert system disclosed in the Karis patent stores data related to a set of previous printing runs or cases in which the problem, e.g., a web break, actually occurred. An expert then goes through the cases and identifies the most likely reason or reasons that the problem occurred in each case. These reasons are then stored in the memory of the expert system and, thereafter, the stored cases, along with the cause and effect reasoning provided by the expert are used to classify the cause(s) of the problem when it arises later. Unfortunately, the Karis system requires the use of an expert to originally identify the most probable cause(s) of the problem and, thus, is totally dependent on the expert's knowledge and reasoning. The Karis system does not identify causes that were never identified by the expert because, for example, the expert did not connect the problem to a particular cause or because the cause did not result in the problem in one of the cases reviewed by the expert. Furthermore, the Karis system does not store or collect data pertaining to printing runs in which the problem did not occur. As a result, the Karis syst
Evans Robert
Wong Did Bun
Marshall Gerstein & Borun
R.R. Donnelley & Sons Company
Starks, Jr. Wilbert L.
LandOfFree
Using ink temperature gain to identify causes of web breaks... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Using ink temperature gain to identify causes of web breaks..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Using ink temperature gain to identify causes of web breaks... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3033759