ISA TR88 95 01-2008 Using ISA-88 and ISA-95 Together《共同采用ISA-88和ISA-95》.pdf

1、TECHNICAL REPORT IS A-TR-88.95.01-2008 Standards Certification Education ISA; 67 Alexander Drive; P. 0. Box 12277; Research Triangle Park, NC 27709; Telephone (919) 549-8411; Fax (919) 549-8288; E-mail: standardsisa.org. The ISA Standards and Practices Department is aware of the growing need for att

2、ention to the metric system of units in general, and the International System of Units (SI) in particular, in the preparation of instrumentation standards. The Department is further aware of the benefits to USA users of ISA standards of incorporating suitable references to the Sl (and the metric sys

3、tem) in their business and professional dealings with other countries. Toward this end, this Department will endeavor to introduce 51-acceptable metric units in all new and revised standards, recommended practices, and technical reports to the greatest extent possible. Standard for Use of the Intern

4、ational System of Units (Sf): The Modern Metric System, published by the American Society for Testing ;:ru1 L I:onmin f“r-rn rn.o-rl l Site L.“T“:onmin ,., . ,.uA . , . ,. . r.:; ! Area U.rFH.rU.o “-“J.J May contain May contain Figure 2- ISA-88 Physical Model Process Control, the Unit Supervision ac

5、tivity provides most of the functionality required to execute procedures and to direct the lower level functionality described in the Process Control activity. Since those procedures often must be performed in multiple units and common resources at the same time in a precise sequence with exact timi

6、ng and because there could be many procedures active at the same time, a great deal of coordination is necessary. The coordination functions and the remainder of needed procedural control functionality are defined in the Process Management control activity. By their nature, the interconnection of th

7、e three primary control activities is viewed in ISA-88 as interactive and time sensitive. For that reason, they are essentially inseparable in any practical implementation where there is significant interaction of closely linked elements (units and common resources) of the manufacturing equipment be

8、ing controlled. While the control activities represent the functionality needed to implement procedural control, it is equipment that is, in fact, controlled. Unlike the more traditional view of process control that treats control as separate, abstract activities that focus primarily on manipulation

9、 of final control elements, procedural control addresses and provides functionality to groupings and assemblages of equipment. ISA-88 defines a physical model (see Figure 2) with a hierarchy of four specifically defined equipment groupings and three higher levels, fairly abstract groupings that are

10、depicted in dashed lines because they are specified as beyond the scope of the ISA-88 standard. - 15- ISA-TR-88.95.01 The three primary control activities discussed above rather naturally align with levels in the ISA-88 physical model. The functionality defined in the Process Control activity contro

11、ls the behavior of the equipment groupings called Control Modules and, occasionally, Equipment Modules. The Unit Supervision control activity provides the capability to execute embedded modular procedural elements (phases) to control the behavior of the equipment. The Process Cell is made up of Unit

12、s and common resources, both made up of Equipment Modules and Control Modules. Control functionality in the Process Cell relies on the Process management control activity to track and direct the units and common resources, to coordinate their interactions and to keep complex procedures sorted out an

13、d working together to make the desired product. As a result, the three primary activities are viewed in ISA-88 as tightly bound to the process cell, as well as the unit and the lower level equipment that make up the process cell. The other three complementary activities are specifically not part of

14、the Process Cell and their functionalities transcend individual process cells. In the ISA-88 models, they are not tightly bound to any specific grouping of equipment and are not clearly related to any particular level in the physical model. However, these functionalities may be likely associated wit

15、h equipment groupings in the Area, Site and Enterprise as shown in the physical model but not likely with the Process Cell. When the first parts of the ISA-88 standard were issued, there was no recognized standard defining functions above the Process Cell. As a result, it was necessary to define the

16、 three higher-level control activities in order to make sense of the functions in the lower level control oriented activities. Process Management cannot function without a recipe and a schedule, both of which are likely to come from above or outside the Process Cell. The recipe, or the basis for it,

17、 is usually generated in some corporate R furthermore, the actual boundaries implemented may not correspond to the boundary between Level 3 and Level 2, as defined in the ISA-95 standard. A key focus of the ISA-95 standard is to model information flow among functional entities while the ISA-88 stand

18、ard focuses on modeling the flow of control among equipment entities. Problems tend to arise in the way these entities are implemented. ISA-88 is much closer to being a physical implementation model than ISA-95. By treating the ISA-95 models as abstractions of physical implementation models and by r

19、ecognizing the ISA-88 implementation models of equipment entities, one can realize the combined use of these two standards. To avoid problems due to overlaps, terms need to be rationalized, operational correspondence clarified and implementation examples given. Fortunately the functional boundaries

20、in the two standards appear to be quite similar and the included functionality appears to be nearly the same. It is only when function is combined with and bound to equipment that confusion is likely. The ISA-88 equipment entity concept and ISA-95 segment concepts need to be recognized and understoo

21、d to improve clarity where the two standards must interact. In particular, extending the much more broadly defined ISA-95 segment concepts from Level 4 to Level 3 (as is commonly done in practice, but not currently defined in the standard) provides a layer that resolves into the ISA-88 recipe entiti

22、es that affect its equipment entities. Both standards, in practice, are not only used for automation purposes, but are also useful to clarify and promote communication about manufacturing concepts, systems and technologies between people and groups. This often leads to the need to choose between ter

23、minology and models. 3.3 Key gaps in models, definitions and terminology In addition to overlaps in the terms, definitions and models, there are gaps between the two standards that also need to be recognized and understood. There are terms, definitions and model elements in ISA-88 that are not inclu

24、ded in ISA-95, as well as, vice-versa. Some of these “gaps“ are mentioned in the detailed tables in Clause 4. A significant gap in the two standards is in the types of production operations being addressed. ISA-95 is designed to apply to several types of production, including batch production, while

25、 ISA-88 was originally intended for batch manufacturing alone; however, ISA-88 can potentially be extended beyond batch to other types of manufacturing as described below. ISA-88 was written in terms of batch manufacturing, even though there was no intention to limit it to batch alone if there are w

26、orthwhile applications elsewhere. The reason for the batch focus is obvious. ISA-88 was written because there was no standard way to address automated procedural control in process manufacturing. This was particularly evident in the case of batch manufacturing and that was the problem the standard a

27、ddressed. Though focused on batch, the standard actually defines an effective and internally consistent way to apply procedural control to many types of manufacturing processes. It is certainly not the only way to apply procedural control, but it has been successfully applied in many cases to many d

28、iverse types of manufacturing. In spite of this, the terminology and details remain a problem. Because it is couched in very batch oriented terms, if one reads the words literally and without interpretation, it can be easily dismissed as a niche standard applying only to batch manufacturing. Increas

29、ing use of ISA-88 in other types of manufacturing, though, has demonstrated that it may be a desirable guide in other kinds of applications and may be the desired approach to automation whether the application is batch or not. The use of ISA-88 in a broader array of manufacturing technologies is bey

30、ond the scope of this technical report. However, to minimize confusion, it is worthwhile to recognize that it may be used in those applications and to understand why and a little bit of how it might show up. Practice has demonstrated that understanding the potential use of ISA-88 approaches in some

31、continuous and discrete applications of projects is needed because few manufacturing processes are 100% pure batch, pure continuous, pure discrete or pure anything. Even within batch manufacturing, for example, the product (the batch) is quite often not a directly saleable product until is packaged

32、or subjected to further processing. Auxiliary manufacturing processes are needed, often as part of the same overall manufacturing facility. There are advantages to using the same general approaches to automation of closely related equipment as long as they are applicable. It is not difficult to visu

33、alize possible approaches utilizing ISA-88 principles in some continuous, discrete and/or storage situations. Continuous processes are a good example. They differ from batch processes in several ways. They tend to have more processing equipment, be more costly and generally do not make very many dif

34、ferent products. They are highly productive in terms of pounds of product per labor-hour. The regulatory control is often very sophisticated and expensive because even a tiny improvement in throughput or yield can, over days or weeks or years, result in large benefits. In many continuous processes t

35、he procedures for startup, shutdown and grade change are manual. However, if continuous processes are examined only in the context of procedural control, most continuous processes can be thought of as batch processes with a long, unchanging in-process step. It is true that to apply ISA-88 to most co

36、ntinuous processes requires that we interpret the intent of the standards batch specific “rules“, but this is often worth the effort. Certainly, startup and shutdown under automatic procedural control is useful in many “continuous“ applications. Since these are the times of greatest uncertainty, wit

37、h the greatest need for extensive and immediate exception handling, procedural control automation may be desirable or necessary. It - 21 - ISA-TR-88.95.01 would appear that expanding the ISA-88 standard to encompass both batch and continuous would not be difficult except for the tedious business of

38、determining what to call things when they have multiple names. If an expanded and broadly accepted view becomes common, ISA-88 principles will become even more prevalent in continuous (and other) manufacturing processes. Discrete processes are more of a challenge, not because the principles are diff

39、erent but because the implementations are so diverse. ISA-88 is being applied to some discrete manufacturing processes including such examples as packaging lines made up of multiple automatic packaging machines. Generally, control modules and equipment modules may be buried in the proprietary machin

40、es, but units are not hard to identify and the entity that controls a procedure to convert a defined quantity of raw material (perhaps potato chips) into a finished or semi-finished product (perhaps bags of chips) is not hard to find. These examples are not an argument to force fit a batch standard

41、to replace either existing systems that work or other approaches that are better. The point is that ISA-88 can be used with some benefit if it is generalized to make clear how it fits the different environments and, for that reason, may be encountered in many different kinds of applications. Though

42、not yet published, Part 5 of !SA 88 is attempting to address some of these very issues. Neither continuous nor discrete processes exist in a vacuum and are frequently intermixed with each other and with batch. While ISA-88 (or a derivative of it) is unlikely to become a universal standard for manufa

43、cturing, having a more or less standard approach to procedural control in all of the manufacturing entities in the same facility could be, in many cases, beneficial and something an individual or team applying ISA-95 may well encounter outside of “pure“ batch applications or projects. Some of these

44、issues have already been considered. The ISA-95 committee has elected to generalize the names of the various entities that are roughly equivalent to the ISA-88 Process Cell and the ISA-88 Unit. ISA-88 Process Cells and their equivalents in other types of manufacturing and storage/retrieval applicati

45、ons are all called ISA-95 Work Centers. Likewise the term ISA-95 Work Unit has been defined as the general term that refers to the unit in a batch process and similar structures in all types of manufacturing. This is a powerful concept. Since the Work Units are likely to be specific to the type of m

46、anufacturing and/or storage, it should be possible to have a continuous oriented work unit as part of a batch process cell (or work center). This solves many problems not addressed by ISA-88. In reality, batch work units are frequently part of predominately continuous work centers. Things that look

47、like storage oriented work units are not strangers in any work center. From a very general standpoint, the mappings and overlaps noted in Figure 7 are similar to those depicted in Figure 6. In those cases where an ISA-95-based application involves a mix of batch, discrete and continuous production c

48、ontrol, the nature of the overlaps in Figure 7 can be different from those noted in Figure 6. Another significant difference in the two standards involves other categories of manufacturing operations management that ISA-95 explicitly addresses in addition to production operations, such as, maintenan

49、ce operations, quality assurance testing operations, and inventory movement operations. The ISA-95 generic activity model shown in Figure 5 can be replicated for each of the manufacturing operations management categories. Additional diagrams similar to Figure 6 can be constructed but with the term “Production“ replaced with either “Maintenance“, “Quality Assurance Testing“ or “Inventory Movement/Storage/Retrieval.“ The ISA-88 models relate directly to the ISA-95 Production Model, however ISA-88 models may also relate to such activities as ISA-95 Inventory Movement