Avi Wasser

Semantic Machine Learning for Business Process Content Generation

Business process modeling is considered a manual, labor intensive task. It requires significant domain expertise and may be prone to errors or inconsistencies due to reliance on human factors. Hence, automation through reuse of predefined process models is becoming a common practice for generating new models. In this work we extend a previously proposed generation method by adding semantic learning capabilities that opt to improve the quality of generated business process models. The learning mechanism analyzes, in real-time, the linguistic relationships between process descriptors and adjusts them according to human inputs that are accumulated during the modeling process. To demonstrate the method we present a case-study from the food manufacturing industry. To estimate the applicative value we further experimented the method on a real-life process repository, showing that the learning mechanism increases the effectiveness of the previously suggested method for automating the design of new business process models.

Ontology Based Method for Supporting Business Process Modeling Decisions

This work suggests a method for machine-assisted support of business process modeling decisions, based on business logic that is extracted from process repositories using a linguistic analysis of the relationships between constructs of process descriptors. The analysis enables the setup of a descriptor space, in which it is possible to analyze business rules and logic. The suggested method aims to assist process designers in modeling decision making, based on knowledge that is encapsulated within existing business process repositories. The method is demonstrated using a real-life process repository from the higher-education industry.

Applications for Business Process Repositories Based on Semantic Standardization

In recent years, researchers have become increasingly interested in developing frameworks and tools for the generation, customization and utilization of business process model content. One of the enabling central techniques for automated repository standardization is Natural Language Processing (NLP). This work reviews previous works on NLP standardization, and presents a set of derived Business Process Management (BPM) applications. We then discuss how these applications can be extended and improved for better utilization of the process repositories by (1) deploying a larger set of semantic models; and (2) integrating complementing applications.

Natural Language Processing for Risk Identification in Business Process Repositories

In recent years, researchers have become increasingly interested in developing methods and tools for automating the design of governance, risk and compliance (GRC) models. This work suggests a method for machine-assisted identification and design of new risks, based on business logic that is extracted from real-life process repositories using a linguistic analysis of the operational similarity between process conducts. The suggested method can assist process analysts, audit executives and risk managers in identifying new organizational risks while making use of knowledge that is encoded in existing process repositories. The suggested framework was tested on the ProcessGene process repository, showing our approach to be effective in enabling the identification and design of new risks within real-life business process models.

Managing Resources in the Generation and Development of Research Based Spinoffs: Evidences from Israeli ICT Cases

Research-based spinoffs (RBSOs) have raised a great interest in recent years for their economic and social implications. The literature has studied this phenomenon from three different perspectives: resource-based, business-model driven, and institutional. Due to the possible correlation between the approaches, this work aims to suggest an integrated framework, adopting the four stages model of spinoff development. To present the application of the framework we performed an extensive examination of two RBSOs operating in the ICT sector. Results highlight the level and types of technological, human, financial and social resources (including also the institutional linkages), and their evolution and importance in the generation and development of these new high tech companies.

Machine-Assisted Generation of Semantic Business Process Models

Generic reference models are based on the assumption of similarity between enterprises - either cross industrial or within a given sector. The research suggests a machine assisted methodology and tools for the design and generation of individualized business process models based on generic repositories. An empirical evaluation is included to assess the effectiveness of the proposed method.

Key Deficiencies of Semantic Business Process Search

In recent years, researchers have become increasingly interested in developing frameworks and tools for searching business process model repositories. While research on searching structured repositories has been extensive, little attention was dedicated to searching business process content within unstructured repositories, such as the Web. We demonstrate why current search technologies are not useful for extracting process content from the Web, and explain the core reasons for the deficiency. We then express the requirements for a framework that could overcome the presented shortcomings.

Selection of Business Process Alternatives Based on Operational Similarity for Multi-subsidiary Organizations

This work suggests a method for machine-assisted support for multi-subsidiary organizations in selecting business process alternatives, based on operational similarity. Operational similarity between processes can be derived from process repositories using a linguistic analysis of process descriptors. The suggested method aims to assist operation managers in multi-subsidiary organizations in identifying similar processes, that can substitute processes that cannot be carried out within a certain subsidiary. This decision making is based on knowledge that is encapsulated within existing business process repositories. The method is demonstrated using a real-life process repository from the paper manufacturing industry.

Decision Support for Operational and Financial Risk Management - The ProcessGene Case Study

This work suggests a generic framework for risk related decision making from a business process management viewpoint. The framework is based on the methodology embedded in the ProcessGene Risk Management software suite. The suggested method aims to assist risk managers in making risk related decisions along the entire lifecycles of risk management, governance and compliance. This decision making is based on knowledge that is encapsulated within existing business process repositories. The method is demonstrated using a real-life process repository from a manufacturing industry.

ProcessGene view: an integrated visualization of business process models

One of the main challenges in the field of business process management is creating visibility into the business of organizations. The prevailing approach utilizes conceptual business process modeling as the foundation for creating and managing this visibility. Our approach aims to extend previous academic and industry efforts in this field by suggesting a systematic business process visualization method - featuring the multi-faceted visualization interfaces to an existing process model. Each interface encapsulates the data in a different manner- aiming to serve different user populations as well as different use-cases. To test and illustrate the suggested method we implemented it as an add-on to an off-the-shelf BPM software suite.