- 06 June, 2003 13:32
Delivering an integrated knowledge environment at Defence.
As with industry, Australian government organisations such as the Department of Defence’s Defence Materiel Organisation (DMO) have recognised the importance of transforming operational data into consistent, accurate and actionable data for the development of good strategy, and timely and informed decision making. Also like strategists in the private sector, they understand that data warehousing technologies, when applied effectively, can be the key.
As organisations demand almost instantaneous access to up-to-date information and analysis, strong data warehouse solutions are becoming an essential piece of an effective information management strategy. However, until relatively recently, organisations have been bound to the limited capabilities of legacy systems that were designed for reporting, not analysis — which in turn led to lengthy, expensive IT development efforts in an attempt to achieve desired analytical capabilities.
The Big Picture
Before looking at the special requirements of the DMO it is helpful to understand how data warehousing fits in the overall IT scheme of things. Industry has recognised the value of an enterprise data warehouse sitting at the centre of an organisation to answer many strategic as well as tactical questions. It is fundamentally different from any other system in that it does not support any day-to-day operational processes and transactions; rather it provides a universal ability to answer key questions about the organisation, particularly the performance and predictability.
A data warehouse focuses on extracting, transforming and loading the best information available at the lowest level of detail from the pre-eminent source systems, providing a distinct location for decision-support enquiries. This provides a different environment from online transactional processing (OLTP) capabilities in the sense that it is designed to manage large data amounts to provide online analytical processing (OLAP) analytical capabilities.
Data warehouses are not designed to replace legacy OLTP systems that are used for operational reporting — OLAP provides the ability via a Web interface to conduct online, drill-down analysis and data mining as well as ad hoc analysis. These capabilities will greatly improve the ability to analyse data to support decision making, making research more effective and efficient. Users of OLAP no longer have to become experts on various legacy systems — and gain the benefit of a “common look-and-feel” of all of the data in the warehouse (regardless of the source system).
The value of a data warehouse is in its ability to quickly support often critical cross-functional business decisions. Decision makers need tools that provide rapid data access and the capability to then perform detailed investigations such as multidimensional analysis and data mining. This will help to determine the best course of corrective or preventive action in the least amount of time. For example, the power of a data warehouse solution can help to identify opportunities to reduce costs and gain efficiencies in inventory management, personnel management, supply chain management and procurement.
The Tempo Increases
Among the DMO’s challenges has been the long cycle time to acquire accurate analytical data and scarcity of analytical capability. At the same time the increase in Operational Tempo for Australian Forces worldwide is driving shorter decision times.
A recent BearingPoint investigation of the DMO reasserted a conclusion made by the DMO Business Information Systems Team in September 2002: there were more than 130 different business functions managing information in more than 25 different subject areas or “themes”, creating more than 3000 logical recording and management points for data in the DMO.
Defence personnel face significant challenges in performing data analysis. A key test is the amount of time currently required to gather accurate data to make critical decisions. Necessary data resides in various, disparate locations . . . in various formats . . . with various methods for access and eventual manipulation.
At present, those requiring the data must become familiar with how to obtain the data from the source system and then present in a usable format — often a lengthy process. Often, individuals would create their own “stores” of duplicated data, and would use various applications and tools to gather and re-format the data, to create reports. By the time this process is complete, the data is often out of date.
The DMO’s ongoing Data Warehouse Initiative (DDWI) aims to provide a capability throughout the DMO (and hopefully across the entire Department of Defence) that brings copies of data extracted from disparate source systems across the DoD into a single environment to enable better modelling and analysis to support the activities of the Australian Defence Forces.
Developed by the Management Information Systems Division (MISD), DDWI will seek to provide an information management environment that enables materiel management excellence and informed decision making across the DoD through the planning, acquisition and in-service support of Materiel Information Systems (MIS) and the stewardship of information contained in those systems.
For both government and commercial entities, successful implementation of data warehousing solutions requires careful project management and delivery, to include:
- proper project scoping early on in the effort to provide appropriate focus, support and funding
- focused project management to monitor and control program performance, cost, schedule and risk
- proactive project execution, to include decision and conflict resolution, change control, life cycle management, communications, knowledge sharing and problem response
- ensuring access to staff with appropriate expertise
- setting and managing user expectations, taking care not to over-promise
- meeting security requirements and access restrictions as data is aggregated
- ensuring access to appropriate technologies in a timely manner
- designing in stability and reliability by applying an iterative, rapid development approach to quickly field incremental delivery of data warehouse capabilities (which in turn will help gain buy-in and provide a demonstrable return on investment)
- adherence to quality assurance through techniques such as Issue Management and appropriate In Process Reviews.
Creating a data warehouse is no simple task. Despite the challenges, the rewards of developing an integrated knowledge environment are well worth the effort.
Paul Nadeau is a managing director at BearingPoint Australia and leader of the company’s Public Sector group