چارچوب جامع برای مدل سازی قواعد کسب و کار مبتنی بر مجموعه در طراحی پایگاه داده های مفهومی

Business rules are the basis of any organization. From an information systems perspective, these business rules function as constraints on a database helping ensure that the structure and content of the real world—sometimes referred to as miniworld—is accurately incorporated into the database. It is important to elicit these rules during the analysis and design stage, since the captured rules are the basis for subsequent development of a business constraints repository. We present a taxonomy for set-based business rules, and describe an overarching framework for modeling rules that constrain the cardinality of sets. The proposed framework results in various types constraints, i.e., attribute, class, participation, projection, co-occurrence, appearance and overlapping, on a semantic model that supports abstractions like classification, generalization/specialization, aggregation and association. We formally define the syntax of our proposed framework in Backus-Naur Form and explicate the semantics using first-order logic. We describe partial ordering in the constraints and define the concept of metaconstraints, which can be used for automatic constraint consistency checking during the design stage itself. We demonstrate the practicality of our approach with a case study and show how our approach to modeling business rules seamlessly integrates into existing database design methodology. Via our proposed framework, we show how explicitly capturing data semantics will help bridge the semantic gap between the real world and its representation in an information system.

Business rules are an important asset of any organization [1]: they represent decisions that are made to achieve enterprise objectives [2] and reflect the business policies of an enterprise [3]. While a business policy is a general statement or direction for an organization, a business rule defines or constrains some aspect of a business, asserts business structure and influences the behavior of an enterprise [4]. From an information systems perspective, these business rules function as constraints on a database helping ensure that the structure and content of the real world—sometimes referred to as miniworld—is incorporated into the database. Business rules need to be captured during the conceptual design stage, as the modeled rules are the basis for subsequent development of a business constraints repository. During conceptual database design, a conceptual model provides a notation and formalism that can be used to construct a high-level description of the real world, referred to as a conceptual schema. However, extant conceptual models provide limited support for set-based business rules, more specifically, rules that constrain the structure or cardinality of sets. These set-based cardinality requirements specified for an enterprise and enforced on a database are also referred to as business constraints in this paper. We present a taxonomy for set-based business rules and describe a framework that is based on various types of conceptual modeling abstractions. We define an overarching framework, which would help elicit set-based cardinality requirements during conceptual design. Further, we explicate the semantics using first-order logic. We show how the proposed framework integrates with extant conceptual design methodology and describe how a formally defined framework can be embedded into an existing design-support environment in a straightforward manner. The ability of an organization to understand and manage its business rules plays a key role in its operational efficiency and market competitiveness [1]. Eliciting business rules helps organizations understand their own operating environment [5]. From a database perspective, business rules dovetail with relational technology [6] as a database may be considered as a collection of propositions. Many conceptual models have been proposed that capture the meaning and structure of the data. However, most of them [7], [8], [9] and [10] capture only a limited range of constraint types. In order to overcome these problems, constraint definition languages and constraint enforcement systems have been developed to provide declarative support for expressing constraints [11], [12], [13] and [14]. However, constraint definition languages are more oriented towards logical database design than conceptual design. Complex constraints expressed in these languages can be hard to understand by users and they may be inadequate for communicating business rules between users and database designers. Prior studies, e.g., [15] and [16], attribute project failures to the lack of identifying real needs during conceptual design. Moreover, the inability to model business rules may even result in inconsistent enforcement [1]. Initially the business rules captured in information systems may be synchronized with the intended policies; however, the subsequent maintenance may introduce inconsistencies resulting in information systems that constrain the ability of the businesses to change. Therefore, there is a need for a formal approach to comprehensively capture business rules at the conceptual design stage. Direct representation of these rules in a database system can be used to ensure that constraints are uniformly enforced for all users [14]. Additionally, these constraints that are embedded in the database can be used for query optimization techniques [17]. Reasoning on embedded constraints has also been used in fields such as semantic query processing and knowledge discovery in databases. Maintenance—estimated to be around 60–80% of project cost [18]—forms the largest expenditure in a typical information systems project. The explicitly captured business rules can help manage system evolution [19], and enhanced system understanding can help ease the maintenance of information systems. Explicit documentation of business rules results in a better understanding of applications resulting in decreased maintenance and testing costs, increased consistency of business practices within and across applications, and improved administration of business rules [5]. Hence, identifying and embedding business constraints in the schema during conceptual database design has ramifications for creation and maintenance of the database system, and the overall operational efficiency and adaptability of an enterprise to a changing business environment. Consequently, prior research [2], [20], [21], [22], [23], [24] and [27] emphasizes the need for eliciting business rules and the ability to dovetail rules with a constraints repository. Rosca et al. [2] describe many benefits of explicitly modeling business rules during conceptual design: (i) helps analyze the completeness and consistency of the operating principles of an enterprise; (ii) enables top-down analysis of how well the enterprise objectives are dovetailed with the rules; (iii) ensures bottom-up analysis of how well the developed applications conform to the operating principles; and (iv) supports more adaptable systems. Martin and Odell [28] classify business rules into constraint rules and derivation rules. While the former define restrictions, the latter define rules by inference from some other rules. In this paper, we describe a formalism and notation to comprehensively capture set-based business constraints of an enterprise. A set is a collection of elements, and our framework focuses on constraints based on the cardinality of a set. In order to capture set-based business constraints during conceptual design, we propose that the database schema developed during conceptual design be augmented using our proposed framework. Our framework includes a grammar (Appendix B) in Backus-Naur Form (BNF) that can be used to capture users’ requirements. We provide precise semantics associated with the framework using first-order logic; these semantics can be the basis for the development of a business constraints repository. We describe a case study that demonstrates the practicality of our approach. In this paper, we apply the business constraints framework to a conceptual model, the Unifying Semantic Model (USM) [29], which is an extended version of the Entity-Relationship (ER) Model [8]. However, our approach is not specific to USM and can be applied to any semantic model, e.g., ER [8], Extended ER [17] and UML [30]. Our work makes several contributions related to capturing business rules during conceptual database design. We present an overarching framework that integrates with and extends prior work related to modeling set-based cardinality requirements. Such a framework plays two roles: firstly, it formalizes some constraints that cannot be captured with any existing conceptual modeling formalism; secondly, it generalizes some existing constraints into its overall formalism. We define seven different types of set-based business constraints on a conceptual model that supports abstractions like classification, generalization/specialization, aggregation and association [31]: attribute, class, participation, projection, appearance, overlap and co-occurrence. In extant models, e.g., [8] and [17], the cardinality and participation constraints [17] apply to interaction relationships only. We show how such constraints can be generalized for abstractions like generalization/specialization, composite and grouping relationships (cf. Section 4.3). The attribute constraint (cf. Section 4.1) generalizes the notion of optional/required attributes, which imply a minimum attribute value of 0 and a maximum attribute value of 1, respectively. With attribute constraint, any number associated with the minimum and maximum attribute value can be specified. Additionally, we have defined set-based constraints like appearance, overlap and co-occurrence as they apply to different types of conceptual modeling abstractions (cf. 4.4, 4.5 and 4.6). From a practical perspective, such a framework will provide a template for capturing set-based cardinality requirements during design and analysis stage. Additionally, formally defined semantics can be incorporated into an existing database design tool. From a researchers’ perspective, we provide an overall framework for defining various types of business constraints on conceptual modeling abstractions. Such a framework can be the basis for future work like extensions along other orthogonal dimensions like time and space. We formally define the syntax in BNF and the semantics using first-order logic. The syntax of the proposed framework is expressed in a form that is straightforward and concise from the perspective of a database analyst. Additionally, the business rules must be represented formally for automated execution [25]. The semantics associated with our framework described using first-order logic can be applied for automated rule execution via assertions. We define the concept of metaconstraints, i.e., constraints on constraints. Metaconstraints may be regarded as an implication from partial ordering among the constraints. A comprehensive classification of constraints and the partial ordering among constraints can be used for automatic consistency checking by a design-support environment during design stage itself. We present a methodology that integrates our framework with the extant conceptual design methodology. To underscore the practical focus of our approach, we conducted a case study. The case study at the admissions office of a large public university demonstrates how the framework can be applied to capture business constraints within the existing conceptual database design methodology. In summary, this paper presents an overarching framework that would help capture set-based cardinality requirements and defines the notion of metaconstraints that would help identify inconsistency in constraint definition even before the database is populated. We describe a formal, intuitive, straightforward and practical to implement approach that integrates with the existing database design methodology. The rest of the paper is organized as follows: in Section 2, we motivate the need for a business constraints framework using two examples. Next, we present a taxonomy for business constraints using two orthogonal dimensions: types of abstraction and categories of constraint. In Section 4, we describe the business constraints framework based on the taxonomy presented in the previous section. Throughout, we use examples introduced in Section 2. The necessity and usefulness of capturing the set-based business constraints is demonstrated with a case study in Section 5. We briefly describe related work in Section 6. In Section 7, we summarize our work and discuss future research directions.

Developing a repository of business constraints has wide implications: it can help in consistent enforcement of business practices across applications, decreased maintenance and testing costs and improved administration of business rules resulting in an overall improved operational efficiency and adaptability of an enterprise to a changing business environment. However, extant conceptual models are incapable of capturing various types of business constraints. In this paper, we established a comprehensive framework for modeling set-based business constraints that can be imposed on different types of conceptual modeling abstractions. Our framework is based on two orthogonal dimensions: types of abstraction and types of constraints. By introducing a unified conceptual framework for set-based business constraints, this paper motivates and helps establish a procedure for capturing and defining business rules at conceptual design stage. As demonstrated by the case study, the business constraints framework can express many business rules that may not be possible using the formalism of traditional conceptual models. Via our proposed framework, we show how explicitly capturing the data semantics will help bridge the semantic gap between the real world and its representation in the information systems. We understand that a completeness proof cannot be given for any conceptual framework since the framework is motivated by practical rather than purely theoretical reasons. New constraints may be proposed and these constraints will help capture more data semantics associated with the real world. However, we believe that our framework can be easily extended because it is based on two orthogonal dimensions. Any extension will either require adding a new dimension or adding a parameter to the existing dimensions. In ongoing research, we are exploring several important aspects. One of the challenges in requirements elicitation is that typically the stakeholders are numerous, distributed and with possibly conflicting goals [50]. We plan to extend the design environment so that multiple partial views of business rules can be specified and the integration of these views can be facilitated. Nuseibeh et al. [20] and [21] describe how heterogeneity of representation poses challenges related to integration of: (i) methods, (ii) tools that support the methods and (iii) multiple specifications fragments produced by employing diverse methods and tools. They present a ViewPoints framework that facilitates encapsulation of partial specification and integration of the components. We believe that metaconstraints can support identification and integration of partial and incomplete specifications. We will explore how to employ ViewPoints framework so as to support efficient and effective collaboration in business rules elicitation. In the future, we also plan to extend the formalism by adding spatial and temporal aspects into the framework. Many real world applications, e.g., accounting, portfolio management, personnel management and inventory management, are temporal in nature [51], [52] and [53]. In other words, these applications need to organize data based on time. Lately, geographic information is increasingly employed in a wide array of applications including social, environmental and economic studies [54], e.g., land information systems, environmental modeling, resource management, transportation planning, geo-marketing, geology and archaeology [55]. Mennecke and Crossland [56] describe how geospatial information can be applied for business applications like facility management, market analysis, transportation, logistics, strategic planning and decision-making. Thus, with recent advances in technologies like high-resolution satellite-borne imaging systems, mobile systems, global positioning systems and overall decrease in hardware costs, temporal and spatial data is finding its way into many traditional applications. A constraints framework that helps elicit temporal and spatial business rules will be critical in the development and maintenance for the up-and-coming applications.