The Event State Action (ECA Rules) approach [37] was developed to address the need to respond to different types of events in active databases [38]. An ECA rule consists of three components: ECA rules can also be used in rule modules that use variants of the rete algorithm for rule processing. Deductive database systems. Deductive database systems [58] combine rules-based logical programming with a relational database so that implicit knowledge can be derived from explicit information stored in the database. In deductive databases, the rules are usually expressed in Datalog [59], a subset of prologue that aims to enable effective evaluation in large databases. In recent years, several mechanisms for self-adaptation have been proposed by various research communities, including the use of graph grammars [11], machine learning [12], control theory [13, 14], intelligent agents [15], event-condition action rules [16, 17], and execution models [18]. Therefore, even for experienced architects, it remains difficult to be informed about all candidate alternative architectures, to make sound decisions about trade-offs, and to conduct an early and rigorous analysis of quality features. Several models of how business rules should be integrated into business processes were introduced by Graml et al. [181]. The rules can be complex, but the result, which is a simple Boolean value (yes/no), can be used to control the process. Similar to the work of Vanderfeesten et al. [115] This approach calls for a rules service to evaluate rules before and/or after an activity in a process.

From an architectural point of view, both share the same concepts of integrating rules. However, this approach has brought some improvements compared to the work of Rosenberg et al. [179]. First, the approach does not limit its adaptability to a simple assessment of the rules. The introduced templates specify several actions that can be performed based on the evaluation of the rules, including even more drastic changes such as inserting subprocesses into the process flow. Second, the rules-based approach allows decisions to be made based on a common process context, which is important when multiple rules are used together to make decisions. Third, BPEL code post-processing is done automatically via XSLT transformations [182, 183][182][183], reducing engineering effort. Rules-based BPMN (rBPMN) [184] is another attempt to integrate rules and processes. The rBPMN approach integrates BPMN processes with Reverse Rule Mark-up Language (R2ML) rules [185]. Adaptability is ensured by dynamically changing rules.

R2ML rules are used as gateways by BPMN and are connected to other parts of a process flow. The rBPMN approach claims to support all previously proposed models of change [181]; However, rBPMN is more advanced because rule integration occurs at the metamodel level, resulting in a definition of a rules-based process modeling language. The approach uses response rules that can generate complete service descriptions. This is generally not supported by approaches that use the rule integration technique. Constraints are used to ensure the integrity of the entire control flow. However, replacing process fragments is limited to pre-identified parts of an entire process flow. To take advantage of these benefits of business rules, Rosenberg et al. [179] propose to integrate business rules into a service orchestration. A service orchestration specified in the BPEL language is integrated into the business rules. Business rules are classified into constraints, derivation rules, and response rules.

Such categorization distinguishes the types of rules that can be used for a service orchestration. Constraints verify the conformity of data, while derivation rules derive information from available knowledge about inference and mathematical calculations. Response rules are ECA rules that perform actions in response to an event in a specific condition. Trigger; Event condition action rules; Responsive rules While business rule languages offer a powerful way to express business policies in a user-friendly way and even in natural languages [122], special attention should be paid to the “how” these rules should be integrated into a process flow. It should be noted that flexibility is not an explicit feature of business rule making. Boyer et al. note that “agility in the development of business rules is not given, (rather) we must build them” [186]. The same goes for integrating business rules with business process modeling and implementation. The separation of adaptation concerns should be done in such a way as to maintain an overall understanding of the process.

In addition, it is also important to ensure modularity and improve reuse. One of the problems with today`s orchestration standards is the imperative or procedural nature of process modeling. Companies operating in more dynamic environments are challenging the paradigm of modeling imperative and procedural processes by failing to successfully grasp the unpredictability and complexity of businesses. Intrinsically, business policies (rules) that are frequently changed are mixed with process logic, resulting in an unmanageable spaghetti process/rule [106]. Examples of active database systems that use eca rules are ACCOOD [39] and Chimera [40]. Expert system techniques are sometimes integrated to allow automatic triggering of rules. In addition to active databases, ECA rules have also been applied in traditional databases, where the condition is a traditional query to the local database, and in storage-based policy engines, where the condition is a test for local data. Event Condition Action (ECA) is the abbreviation for the structure of active rules in active event architectures and database systems. ECA rules are used in active databases to support reactive behavior and were first proposed in the HiPAC project [2]. Damianou et al. [23] claim that Lucent used a Policy Definition Language (pdL), similar to Ponder, to program Lucent switching products. PDL “uses the Event-Condition-Action (ECA) rule paradigm of active databases to define a policy as a function that maps a series of events to a series of actions.” This approach is interesting because, in addition to policy rules, there are policy-defined event suggestions that allow groups of simple events to trigger more complex events.

EPS that apply the ecA rules in the corresponding EPA include Amit [41] and InfoSphere Streams [16]. These systems support event algebra operators analogous to those provided by active database event algebras, such as Snoop [9], where complex expressions can be created with operators such as And, Or, Sequence, etc. to describe the types of event patterns that can be applied to real-time event streams. Two typical uses of ECA rules in analyzing event data include detecting and responding to the appearance of certain types of event patterns in the database that can compromise the integrity of the data in real time, and running business logic for incoming event flows. The SpEM authors defined a model for creating and executing clinical guidelines. Initially, a general-purpose language called PLAN was adapted to provide clinical guidelines. This language follows a rules-based event-condition-action (ECA) approach [76]. This Court control mechanism is mapped into an existing database management system (DBMS), i.e. at the end that proceeds to the adoption of a specified policy by increasing and managing various triggers [75].

The runtime embedded in a DBMS uses these rules to start a policy and activate a specific task based on the events raised. An ECA rule consists of three elements: (a) an event part that contains a so-called transition predicate that lists all possible events that are relevant to the rule (it represents the situation that the rule must monitor), (b) a condition part, which can be any predicate, and (c) an action part, which is a list of executable functions. SpEM defines the following primitives required in a guideline or protocol presentation template: (a) an action that represents any clinical or administrative task to be performed, maintained, or avoided during the policy enforcement process, (b) a decision, that is, a decision, that is, a decision. a selection from a set of alternatives based on predefined criteria in a guideline, (c) a patient condition that is a materialization of the clinical condition of a treated person based on the actions taken and decisions made, and (d) an execution state that represents a description of the current state of a guideline [73].