Integra(ng Process and Data Management: Finally? Marlon Dumas University of Tartu , Estonia Joint work with Rick Hull & the ACSI Project Team 1 KiBP’2012, Rome, 15 June 2012 M.C. Escher
Build‐to‐Order Process Research each line item to find a supplier Customer Assembled Delivered! Good Supplier 1 Customer PO Material PO 1 […] • Delivered! Assembler Material PO n • 1. Customer produces “Customer PO” Supplier n • 2. Research on each line item (work order) • 3. Work orders partitioned into “Material POs” • 4. Suppliers deliver the materials Customer PO may be • • 5. Products are assembled cancelled at any (me Material PO may be • • 6. Products are shipped “rejected” at any (me 2
Tradi(onal Process Decomposi(on Decompose Manage Manage supplier customer Material Assemble Ship cancella(on PO POs 3
High‐Level BPMN Model 4
Data Model & Distribu(on 5
Great, now let’s glue it together… Decompose Manage Manage supplier customer Material Assemble Ship cancella(on PO POs Work Materials Customer Suppliers’ Customers Orders PO’s PO’s Catalogues + Execution status tracked by assembler’s BPMS + Execution status of processes at each supplier 6 Acks to Lior Limonad
Cancella(on penalty for each work order W for each material order M in W if M has been shipped add returnCost(M) to penalty 7
Tradeoff Data-driven analysis How do we get here? Process-driven analysis 8 Acks to Lior Limonad
SOA to the rescue (Erl 2005) So many of them? + public services Process‐ Centric + logic-centric services Service + utility services Task Service Task Service En(ty Service En(ty Service En(ty Service 9
ESB to the rescue… Process‐ Task Service Task Service Task Service centric Service Enterprise Service Bus (ESB) Logic‐centric En(ty Service En(ty Service En(ty Service U(lity Service Service 10
Where’s the catch? ESB ESB: Erroneous Spaghetti Box? 11 Jim Webber 2009, Guerrilla SOA, QCon
What went wrong? One conceptual model for rules If high priority customer and paying by AMEX and policies permit ship before payment Another conceptual model for analytics and dashboards Core BP model is Process Modeling based on activity flows Data Modeling Data is often modeled separately, not related to other conceptual models Lack of coherence adds substantial complexity 12 to already complex systems Slide by Rick Hull et al.
Didn’t we think about this before? • C.J. Rosenquist. “En(ty Life Cycle Models and their Applicability to Informa(on Systems Development Life Cycles”. The Computer Journal 25(3), 1982. • M. Jackson. System Development. Pren(ce‐Hall, 1983 Link from high-levels models to executable models was not developed 13 Acks to Jorge Sanz (IBM)
Pieces of the puzzle Object‐centric BPM • FlowConnect’s Business Object Model • SAP’s “Business by Design” Status‐and‐Ac(on Management • IBM’s Object Lifecycle Explorer • Case management (e.g. FLOWer) Document workflows and “ac(ve” documents • PDF workflows, document rou(ng • Ac(veXML Collabora(on hubs • EasyChair, ConfHub 14
FlowConnect Business Object Model 15 Redding et al. 2008
Ac(veXML <newspaper xmlns=hjp://lemonde.fr …> <(tle>Le Monde</(tle> <date>…</date> <edi(on>Paris</edi(on> <weather> % service call < axml:call service=”forecast@weather.com” > <city>Paris</city> <unit>Celsius</unit> < /axml:call> </weather> Abiteboul et al. 2008 16
Collabora(on hubs (EasyChair) PC members BPM 2012 Track chairs Authors PC chairs PC members KiBP 2012 Authors PC Chairs “Supportive” rather than “Controlling” Can this become an alternative model to traditional service orchestration? 17 Slide by Rick Hull et al.
Assembling the puzzle Data Virtualiza8on • intensional data linked to services Ar8facts Service Interopera8on Hub • Business‐meaningful • Manages service collabora,ons • Encapsulate data & • Observes messages from opera(ons par(cipa(ng services • Non‐trivial lifecycles • Maps messages to data & events Ar8fact‐Centric Service Interopera8on Hub (ACSI Hub) 18
Ar(facts Created Draft Offered Signed Active Completed Lifecycle model Failed Expired Lost Early End (FSM) . . . . . . . . . Info model (Nested data model) Default Customer Asset info Credit ra(ng, Acquisi(on Drao, Final Payment info info etc. info proposals history • Insight: Gives business managers a unified, end-to-end view of their business operations • Communication: Numerous stakeholders have a common basis for understanding 19 • Actionable: Natural mapping to organization & IT levels Slide by Rick Hull et al.
Build-to-Order Artifacts O P e r m o s t u C Work Order Assembly Material PO Manages sub- assemblies and final assemby Handles a Manages work Customer PO, order spawned Manages a Material from creation to for a customer PO, from launch to delivery PO line item delivery (or “rejected”) 20
Ar(fact Service Container (ASC) Handler for queries Clock Supported (incoming) against public artifacts/ handler service call handler Human attributes Interaction Handler Performers, teams Customer PO Human Task mgmt Work Order Material PO Build to Order ASC ASC insulates the artifacts from their environment & maintains intensional Internal Consumed artifact data Artifact Automated Handler Service Calls (outbound) service Types 21 Task Handler call handler
ACSI Hub service participants binding ACSI Hub P1 P2 … Ar(fact … Containers Pn 22
ESB Solu(on (SOA Hub) Receive Customer PO Send CPO ship date Send CPO shipment Manufacturer Receive Cancel CPO Customer Send Materials PO Receive MPO ship date Send Customer PO Receive MPO shipment Receive CPO ship date Orchestration Send Cancel MPO Receive CPO shipment Hub Receive Supplier abort MPO Cust. Cancel CPO Receive Materials PO Send MPO ship date Send MPO shipment Supplier Receive Cancel MPO Send Supplier abort MPO • SOA hub (e.g. BPEL-based) aggregates data and forwards messages to appropriate services 23
ACSI Hub Subscribe to new Customer PO Write CPO ship date Manufacturer Update CPO because shipped Subscribe to Cust. Cancel CPO Customer Create/initialize Materials PO ACSI Hub Subscribe to MPO ship date Create Customer PO Subscribe to MPO shipment Customer PO Subscribe to CPO ship date Write Cancel MPO Line Item Subscribe to CPO shipment Subscribe to Supplier abort MPO Cust. Cancel CPO Material PO Subscribe to new Materials PO Update MPO ship date Artifact schema holds data models and lifecycle models Update MPO shipment Supplier Subscribe to Cancel MPO Artifact instances hold data about process status Write Supplier abort MPO Participating services: Manipulate artifact instances, and Subscribe to changes in artifact instances 24
Scenario: Adding cancella(on Receive Customer PO Send CPO ship date Send CPO shipment Manufacturer Receive Cancel CPO Customer Return cancel CPO penalty Send Materials PO SOA Receive MPO ship date Send Customer PO Orchestration Receive MPO shipment Receive CPO ship date Send Cancel MPO Receive CPO shipment Hub Receive Supplier abort MPO Cust. Cancel CPO Request MPO cancel penalty Request CPO cancel penalty Receive Materials PO Send MPO ship date Supplier Send MPO shipment Receive Cancel MPO Send Supplier abort MPO Return cancel MPO penalty Supplier needs to support penalty method We have to explicityly loop through all relevant Supplier penalties 25
Scenario: Adding cancella(on Subscribe to new Customer PO Write CPO ship date Update CPO because shipped Manufacturer Subscribe to Cust. Cancel CPO Customer Create/initialize Materials PO ACSI Hub Subscribe to MPO ship date Customer PO CreateCustomer PO Subscribe to MPO shipment Subscribe to CPO ship date Write MPO shipment received Work Order Subscribe to CPO shipment Write Cancel MPO Material PO Cust. Cancel CPO Subscribe to Supplier abort MPO Request CPO cancel penalty Subscribe to new Materials PO Update MPO ship date Supplier Update MPO shipment Subscribe to Cancel MPO Send Supplier abort MPO Consider “Request CPO cancel penalty” from Customer Answer is computable as query against artifacts, involving extensional and intensional data 26
Recommend
More recommend
