INSTITUTE OF INFORMATION SYSTEMS Semantic Normalization and Matching of Business Dependency Models 18 th IEEE Conference on Business Informatics Alexander Motzek ∗ Ralf Möller ∗ ∗ Universität zu Lübeck Institute of Information Systems Ratzeburger Allee 160, 23562 Lübeck, Germany {motzek,moeller}@ifis.uni-luebeck.de August, 29 th 2015 MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Motivation ▸ companies becomes more and more connected . ▸ companies depend on their infrastructure . ▸ ultimate goal: assure company’s success MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Motivation ▸ companies becomes more and more connected . ▸ companies depend on their infrastructure . ▸ ultimate goal: assure company’s success - but how? MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Infrastructure = Company? ▸ complex infrastructure dependencies . ▸ this is a real world model from an exercise. ▸ theorem: this is the company . → infrastructure success==company success. MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Assuring success of a company ▸ something fails or is attacked . → local impact. ▸ global impact on company? MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Assuring success of a company ▸ something fails or is attacked . → local impact. ▸ global impact on company? ▸ might even spread ... MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Assuring success of a company ▸ something fails or is attacked . → local impact. ▸ global impact on company? ▸ might even spread ... ▸ ...to dependent nodes... MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Assuring success of a company ▸ something fails or is attacked . → local impact. ▸ global impact on company? ▸ might even spread ... ▸ ...to dependent nodes... ▸ ...to dependent nodes... MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Assuring success of a company ▸ something fails or is attacked . → local impact. ▸ global impact on company? ▸ might even spread ... ▸ ...to dependent nodes... ▸ ...to dependent nodes... ▸ until everything is impacted. MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Assuring success of a company? ▸ yes, there is a global impact on the company! ▸ should one defend? ▸ remove source of evil? ▸ may be worse... MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Assuring success of a company? ▸ yes, there is a global impact on the company! ▸ should one defend? ▸ remove source of evil? ▸ may be worse... ▸ coffeemachine or nuclear control server ? coffeemachine-factory or power plant? MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Infrastructure ≠ Company ▸ a company is more than an infrastructure. � ▸ some are better than others. ▸ business critical resources . MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Business Dependency Model A B C D ▸ business critical devices provide services, i.e., business functions , supporting accomplishment of BF 3 BF 4 BF 1 BF 2 business processes . ▸ two perspectives on company BP 1 BP 2 ▸ goal: understand the company understand the infrastructure CM 1 protect the company MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Understanding the Company ▸ how to acquire this knowledge? ▸ someone who understands the big fussy ball does not know this. ▸ someone who understands business processes does not understand the big ball. ▸ different expertise, different experts . ▸ bridging knowledge gap through probabilistic graphical model . Motzek et al. (2015) and are creatable directly and independently and by experts. ▸ Great! ...? MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Business Dependency Models ▸ obtained directly from multiple experts . A B B A C D BF 1 BF 2 BF 2 BF 3 BF 4 ▸ automatically extracted from multiple BPMN s. ▸ reconstructed from literal descriptions . BP 1 BP 2 ▸ Local views, multiple models. CM 1 CM 1 Lorem ipsum dolor sit amet, consetetur sadip- scing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam non- umy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea. MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Business Dependency Models ▸ obtained directly from multiple experts . A B B A C D BF 1 BF 2 BF 2 BF 3 BF 4 ▸ automatically extracted from multiple BPMN s. ▸ reconstructed from literal descriptions . BP 1 BP 2 ▸ Local views, multiple models. CM 1 CM 1 ▸ But... one consistent model required. → match & merge. Lorem ipsum dolor sit amet, consetetur sadip- scing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam non- umy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea. MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Business Dependency Models ▸ obtained directly from multiple experts . A B B A C D BF 1 BF 2 BF 2 BF 3 BF 4 ▸ automatically extracted from multiple BPMN s. ▸ reconstructed from literal descriptions . BP 1 BP 2 ▸ Local views, multiple models. CM 1 CM 1 ▸ But... one consistent model required. → match & merge. Lorem ipsum dolor sit amet, consetetur sadip- scing elitr, sed diam nonumy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea rebum. Stet clita kasd gubergren, no sea takimata sanctus est Lorem ipsum dolor sit amet. Lorem ipsum dolor sit amet, consetetur sadipscing elitr, sed diam non- ▸ classical business process or ontology matching umy eirmod tempor invidunt ut labore et dolore magna aliquyam erat, sed diam voluptua. At vero eos et accusam et justo duo dolores et ea. problem? MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Classical Problem? = different nomenclatures , languages , abbreviations ▸ ▸ English & Italian ▸ High Level Voltage Control and Distribution = HCD = HLV CD ▸ xuel ≠ muel ≠ ferp ≠ mferp ▸ dorete = 192.18.210.7 = 02-00-D0-12-C7-93 = 718be323-9d58-4ada-9629-81a6f42a9703 → linguistic approaches of no avail . ▸ flow ⇔ dependencies MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Business Dependency Model Merging and Normalization same dependencies, same entity. = sub- graph isomorphism problem (but much easier) DAG, labeled leaves, known topo-order. ▸ exploit dependency structures . ▸ exploit references by external sources (inventories) ▸ match, normalize & merge (almost) linear in number of nodes MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Business Dependency Model Merging and Normalization ▸ we build structure identifying coding e.g., 2- 1-A 3-ABC ▸ naturally sorted , inexpensive lookup ▸ allows for partial matching matches by DEF: F → G , +Q , CD ABC DEF Q ABC CDE DEG CD:+E ▸ sequence alignment ▸ postponed linguistic sufficiency checking MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Simple Example known ⃗ BF : BF q = 02-X Y BF r = 03-V Y Z BF s = 03-X Y Z known ⃗ BP : BP h = 02- 02-X Y 03-V Y Z BP g = 02- 03-X Y Z 03-V Y Z A B BF 1 BF 2 BP 1 CM 1 MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
INSTITUTE OF INFORMATION SYSTEMS Simple Example known ⃗ BF : BF q = 02-X Y BF r = 03-V Y Z BF s = 03-X Y Z known ⃗ BP : BP h = 02- 02-X Y 03-V Y Z BP g = 02- 03-X Y Z 03-V Y Z A B 1. encode CM 1 BF 1 BF 2 BP 1 CM 1 MOTZEK ET AL. SEMANTIC NORMALIZATION AND MATCHING OF BUSINESS DEPENDENCY MODELS, CBI’16
Recommend
More recommend
