AGILE DATA CONVERSION - VALE (MINING)
Vale is the second biggest mining company in the world and the largest in the Americas. Vale is the leading producer of iron ore and pellets and the second biggest nickel producer, and is one of the largest producers of manganese and ferroalloys. Vale’s Information Management infrastructure and processes are vital to the company’s continued ability to monitor and manage its mines, refineries, mills, smelters and operations centers that are located in 30 countries.
Commensurate with its goal of improving its Information Management capabilities, Vale continues to strive to integrate its data and provide true global governance of its information assets. The One Vale Program consists of multiple regional projects to implement global, enterprise-wide business processes supported by the SAP application. The SAP platform is comprehensive, consisting of modules for Finance (FI), Controlling (CO), Project System (PS), Funds Management (FM), Production Planning (PP), Quality Management (QM), Sales and Distribution (SD), Plant Maintenance (PM), Materials Management (MM), Warehouse Management (WM), and Transportation (TD).
Adastra was engaged to analyze, integrate, and transform data into SAP format for the One Vale –Wave 3 Project, the North American and European arm of the One Vale Program. The challenge was particularly complex due to the large number of legacy systems and divisions, holdovers from various acquisitions and mergers.
An important innovation for this IM project was the use of an iterative development approach, which was used for the requirements gathering, development, and QA testing of the extract-transform data conversion programs. This approach provided flexibility and responsiveness when requirements were not yet fully defined, and ensured the ultimate quality of the deliverables. For embracing the innovation of an agile approach for a complex IM project, Vale was awarded a 2013 Adastra Information Management Award (AIMA).
The Adastra team worked closely with Vale to deliver a quality implementation for the target go-live date. From the outset, Adastra applied its combined business and IM expertise to help the business gather detailed and actionable requirements. In addition to technical specifications, Adastra provided data architecture and made sure that the integration was part of a logical, cost-efficient solution in line with Vale’s global vision. Adastra team members became bridges between the various teams and groups whose data was involved in the project; the team identified gaps, and supported the business in filling them.
Over the course of the project, the Adastra team developed, tested, and delivered more than 400 load files ready for input into SAP. Three dry-run cycles were instrumental in fine-tuning the solution to produce data that exceeded expectations in both completeness and quality, allowing for more thorough user testing in SAP. Each dry-run provided the opportunity to refine the requirements and fill newly discovered gaps, resulting in a final production load that was extremely smooth and successful: almost all of Adastra’s delivered files were loaded completely error free and on or ahead of schedule. This was the most successful data conversion project in Vale’s history.
The mandated tool for QA was Informatica DVO. The Adastra team leveraged our Informatica partnership to update the team’s knowledge of the latest version of the tool, and applied Adastra’s comprehensive QA methodology to provide the responsiveness, flexibility, and attention to detail required for any successful QA process. Adastra QA consistently delivered quality results within the project’s tight schedule. In addition, the quality of the documentation produced during QA was such that it was used for internal and external audits.
Data Quality Management
Adastra attaches great importance to Data Quality Management during the Data Conversion process of any project. For One Vale, a Master Data Quality Framework was created to examine the data across six key areas: Completeness, Conformity, Consistency, Accuracy, Duplication and Integrity. Before the start of the project, Vendor, Customer, and Material data was stored and referred to in different ways throughout the organization, resulting in significant duplication as well as conflicting data. The Adastra team supported the business during the requirements phase and ensured that master data was integrated and capable of producing a single version of the truth, which is vital to managing large global organizations such as Vale.
- Budget: $7.7 million for extract-transform development and QA
- Resources: Adastra provided 34 resources to support implementation including Project Delivery Lead, Solution Architect, Business Analysts, ETL Developers, DQ Developers and QA Analysts.
- Duration: 22 months
- Environment: Informatica, SAP
- Notes: Adastra delivered over 400 transformations.