2013 OTM EU SIG evolv applications Data Management
0 likes1,882 views
This document discusses the history of Oracle Transportation Management (OTM) implementation processes in Europe and outlines best practices for data management and user access management. It describes how early OTM implementations relied on individual efforts which led to inconsistencies. As the user base grew, common tools and processes were developed but still varied between projects. The document advocates defining standardized practices to improve consistency, supportability and efficiency across implementations. It provides recommendations for best practices in loading reference data, managing data changes over time, and provisioning user access roles and privileges in a centralized manner.
2013 OTM EU SIG evolv applications Data Management
- 1. Achieving Maturity in OTM Implementation Processes – Part 1 Best practice definitions for Data Management and User Access Management
- 2. History § Oct 2005 (just before Oracle acquired GLog) – under 20 OTM implementors within Europe • Success in Projects achieved through individual endeavours • Regular issues with deployment to producHon – typos in agent code, missing master data etc • Regular issues with user provisioning, eg Hckets raised by business as user in Poland is seeing currency in USD etc
- 3. History continued § 2006 thru 2009 (Steady growth of OTM install base within Europe) • “Home grown” tools emerge built by individual implementors to solve their needs at a parHcular point in Hme (excel funcHons, excel with vba, access db, pl/sql, vb, java applet) • Lack of consistency in approach between implementaHons • Clients encounter performance problems leading to learnings about do’s and don’ts -‐> fed back through development to become the “Performance tuning guide • Projects delivered successfully but: § risk of “home grown” soluHons become unsupportable when the originaHng implementor leaves the project § resources have to learn new processes when moving from one project to another
- 4. History continued § 2010 onward (Rapid growth of OTM install base within Europe) • Conscious effort to define best pracHces • Aim to have consistent approach across all of our new projects • New breed of clients purchasing OTM who previously developed so^ware in house. -‐> Strict code management and deployment processes in place which have to be met. • We start to migrate “home grown” tools to common enterprise grade producHsed plaaorm § Result is that we have now got defined best pracHces and are in the process of rolling those out across Mavenwire.
- 5. Data Loader Rate Maintenance Challenges How to get the rates from Excel to OTM
- 6. Data Management Best Practices § Remove the IT middleman from the day to day process and allow the business end user who is responsible for the data to also load the data • To do this the process must not require the person transforming the raw data and uploading it to OTM to have any knowledge of scripHng/macro/ coding languages • The process should be efficient to use (minimal number of mouse clicks etc) § Use robust repeatable process • TransformaHon of the raw data must produce idenHcal results every Hme • Define consistent naming convenHons • Validate the data is correct • IdenHfy data formaeng issues • Avoid recurrent errors • Verify that foreign keys exist in OTM already
- 7. Data Management Best Practices § The transformaHon must be able to evolve over Hme • Adapt to new opportuniHes (e.g. modes, services) • Simple way for IT to handle change requests to the transformaHon (e.g. new modes, services) • Naming convenHon change should cascade to child objects with minimal effort § Data should be Consistent across environments • e.g. new locaHons loaded to Prod should also be loaded efficiently to test environments § The end user should be able to download the exisHng data, make modificaHons and reload the new data § When new versions of rates are loaded, the old version should be automaHcally expired (this should not rely on the naming convenHon)
- 8. User Access Mgmt Best Practices § Define one consistent model for user access configuraHon § “Level” and VPD profile are for the funcHonal role § VPD context is for the geographical element § User role brings together the funcHonal role and the geographical element. § User provisioning should require creaHon of a gl_user record and assignment of one or more user roles and absolutely nothing else. § AcHon checks, acHon morgs, menus, status type filters, default finder sets are funcHonal role related so should be agached to the “level” § Timezones, currency, business monitor and therefore preference, along with saved query filters are geographical based and should be agached to the user role.
- 9. Data Loader Rates maintenance Options BUSINESS NEEDS § Through the UI? New Contracts • MulHple objects/tabs Contract • Time consuming Expiry § IntegraHon? Rate Updates • Disparate source data Scheduled • No Rate interface Adjustments (e.g. Fuel) § CSV file upload? Preferred Carriers • How do you generate the csvs for many tables • How is it supported? • How is it managed? • Where’s the audit trail?
- 10. User Access Mgmt Options BUSINESS NEEDS § Through the UI? • MulHple objects/tabs Role out • Time consuming functionality to new • Edit User Access screen geographical only available to region “ADMIN” Level § IntegraHon? Add new action check to • No interface all users § CSV file upload? • How do you generate the csvs for many tables • How is it supported? • How is it managed? • Where’s the audit trail?
- 11. How are we internally implementing these Best Practices?
- 13. Data Loader User Access Facilitate the business to load their own data Manager efficiently and robustly Facilitate rapid and robust user provisioning Data Archiver & Cleaner Configuration Meet legal requirements for long term storage of critical Manager data while keeping your Robust deployment primary database running processes efficiently with minimal live data Deliver Support tool for issue, request, feature and bug tracking
- 14. Data Loader Data Loader § Supports multiple environments with different OTM version § User Friendly error handling § Miscellaneous validation & conversion rules § Data can be pushed directly into OTM or loaded using csvutil zip file through the OTM UI § Very quick and robust process § UI for IT personnel to easily manage and create new templates § Extract Data from OTM back into Excel § User Security & Help
- 15. Data Loader Data Loader Templates Template for OTM csv for CE TL RATES CE TL RATES Template for OTM csv for NE RAIL RATES NE RAIL RATES Template for OTM csv for LOCATIONS LOCATIONS § Validation Rules § Mapping Rules § Versioning Rules
- 16. User Access Manager User Access Manager § Supports multiple environments with different OTM version § Locking Mechanisms to force serialisation of tasks if they collide. § Saved queries built from templates plus parameters § Uniform naming conventions enforced across all geographies § Time to create action checks greatly reduced § Load to OTM directly or using csvutil zip file
- 17. Data Loader Data Loader Process § Start with an empty (or pre-‐populated) spreadsheet • Send to Pricing Team • Send to Carrier § Upload spreadsheet • Validate 2 • Error Handling § Load into OTM • csv zipfile 3 • or direct db 1
- 18. Data Loader Demonstration
- 19. Data Loader Other Uses § Load test data into OTM for OTM testing before integration feeds have been built § Load performance test data - eg > 100000 orders with related shipments etc ready for inbound and outbound message testing § Extract business objects from production and load them into a test environment for issue replication § Prototyping § Load test data onto integration stub tables
- 20. Questions ? Questions and Answer Email: [email protected]