MDBS Case 8

Case 8: Data warehousing and multi-dimensional data modelling
Introduction and case facts

ACME Fleet Inc. is a fleet organization that is growing rapidly and has even found its way into
the international markets. The company is currently dealing with huge amount of data due to
the high number of customers who are using or interested in using their services. The company
uses Web analytics and operational systems to gather information about their customers. The
company is currently facing the problem of big data limitations whereby it is unable to make
timely decisions from its data, and still, some information cannot be understood. The company
hence requires a data warehouse which will enable data to be extracted on a periodic basis from
the systems and also be stored in a dedicated server containing the data warehouse. Having a
well-designed data warehouse will allow the company to store its data in a cleaned, formatted,
reorganized, summarized, validated, and supplemented with data from various different
sources. The new Datawarehouse will be the main source of very important information
required for report generation and analysis of different reporting tools that can be used for
certain things such as canned reports, ad-hoc queries, and dashboards. Building a data
warehouse for any organization has never been simple until now where we have the tools for
these processes improved. They make it simpler for organizations to understand the
architectures. In addition to that, there are also some consultants who are always willing to help
at a fee.
Mehta had more than 15 years of experience successfully building and managing a
number of IT projects. Projects were more challenging to execute than normal IT projects and
required organization wide commitment and support. These activities were coordinated by a
store manager and executed by sales assistants and other employees.
Mehta and his team developed a strategy to fulfill the strategic needs of the organization. While
the team believed that the best way to meet those requirements was a data warehouse, several
questions lingered in their minds. What were the possible pitfalls in the integration of a
warehouse system with existing IT? What were the structural decisions to be made when
meeting all the empirical requirements? How do they go about estimating the hardware
resources they use to make decisions? Which were the design alternatives available and which
choice would better fit their needs? What Change Management Initiatives and Communication
Planning will be needed at an organizational level for effective implementation? The company
built a network of regional warehouses to cater to the individual store requirements in the
region. The primary warehouse activities included storage management and distribution,
inventory management, liaising with raw material suppliers and manufacturers for item
procurement, and liaising with transporters for proper and on-time delivery of shipments.
As part of its distribution activities, each warehouse shipped items to a set of stores within its
region. While basic sales data was available, there was no way to get an integrated view of sales
across multiple stores, slice and dice data, conduct multi-dimensional data analysis or build data
mining models to identify the right customer targets. Information Technology Systems at Acme
Inc. Historically, the company had grown through acquisitions.
A data warehouse organized and stored data in a way that was best suited to perform analytical
queries. As per Gartner analyst estimates, the overall market for data warehouse systems alone
was estimated to be about US$ 9 billion in 2013 and expected to grow at a CAGR of 7%.3 The
size of the market, its continued growth and industry success stories clearly demonstrated the
value that organizations could derive from such infrastructural IT investments. Warehouse users
wanted to know the time taken to move a product from the warehouse to the store. Mehta
carefully contemplated the above questions. He knew the answers would be critical in arranging
a pilot study, preparing a detailed implementation plan and presenting the findings to Barley
and other key stakeholders.
Problem Statement
The Acme Warehouse Company received a consignment of 20' lengths of 3" diameter stainless
steel pipes. Acme had never handled pipe as part of their warehousing operation. The forklift
truck operator assigned to this job used the truck's forks as a ram to load, handle, and unload the
pipe. Inspection of the pipe by the owners revealed that the forks were bending and damaging
the pipe. The dilemma is now how ACME can eliminate the problem of pipe damage. From a
cost and ease in application standpoint.
Criteria for evaluation

1) Effective business decisions
The process of 'Data warehousing' helps in improving the speed and the efficiency of
analyzing different data sets and further simplifies the tasks of corporate decision-makers by
helping them to conclude the business insights. Thus, this sound decision making process
help the company to stand above their competitors in the industry.
2) Data consistency
Data warehouses are designed and structured in a way that it helps in applying a
standardized format or layout to all collected data. This procedure helps the management
and corporate decision-makers to analyze, evaluate and share data insights with their team
members and all the related stakeholders. Also, the uniform data from several sources,
reduces the danger of mistake in its interpretation. Thus, it will help the company to
improve overall level of accuracy.
3) Overall profit and the bottom line of company
The platforms of 'Data warehouse' enable the corporate managers and leaders to promptly
prioritize their organization's functional activities and analyze the endeavors that have been
successful — or unsuccessful — in the past.
This helps them to identify the gap and how they can modify or diversify their current
strategy to fill the chasm. Hence, it maximizes the functional efficiency and boost sales
volume which ultimately leads to positive impact on the bottom line of company.
Recommendations
Having understood what a data warehouse is and what should go along with data
warehousing, the company needs to adhere to some of the best practices of data warehousing.
Some of these practices include:
1) Come up with a visual architecture of the data warehouse

During the interaction with the team i.e., at the beginning of the project, the CIO
should come up with an initial architecture model to give the team members an idea
of how the data warehouse will be built. As suggested by Agile Model Driven
Development (AMDD) it is not a must for a model to be so comprehensive or
detailed. It only needs to be a high-level vision model as more details will be
identified vial model storming on a just-in-time (JIT) basis. This initial architecture
will be sketched on a simple whiteboard and its views will be in the form of a
deployment diagram capturing various technologies that will be used and a high-level
domain model showing the business entities and how they are related.
2) Model the details just in time (JIT)

It is a bad idea to model details at the beginning of a project. For this project, model
storming or the model details will be carried out throughout the project in a just in time
manner. The main reason is this is that we assume that some requirements might be
changed hence modelling everything upfront results to a significant wastage of time
and resources.
3) Prove the architecture early

Everything works can work on a whiteboard, PowerPoint slides, and/or in Case tool
models but the architecture will only be proved to be working when proven with
codes. As most processes suggest, the project will be built on a working, end-to-end
aspect to prove that each aspect of the project is working. Since this is a Data
Warehouse project, it will be shown how major legacy data sources can be accessed
and everything else is working including the database regression testing, extra-
transform-load (ETL) strategy and the reporting tools are able to access the DW.
4) Focus on usage
In order to come up with an effective DW/BI system, it is always important to
understand how the users will be using it to support their business objectives. This
means that the project will be a usage-cantered approach driven by usage scenarios
or use cases and not data-centered. This is because focusing mostly on the data
may run us to the risk of implementing something that nobody will be in using
hence a wasted effort.
5) Avoid being hung up on "the one truth"
As the philosophy “one truth” says, it is always desirable has one definition for a data
element or business term. This project will rely on a common definition for the master
reference data or business entities.
6) Organize your work by requirements
Following the agile projects which are based on prioritized requirements rather than
technical issues e.g., source systems, this project will involve working to fulfil the
highest priority stakeholder requirements based on each iteration. Data will be added
little by little in each iteration until the iterations are two weeks in length when a
two-week worth of work is pulled from the top priority stack. Working in this
manner will put the project team in position to achieve maximum benefit for its
stakeholders hence reducing some potential risks.
7) Active stakeholder participation
Just as any other project requires each stakeholder to participate throughout out the
project, stakeholder involvement will be critical throughout the project and
each stakeholder will not only be involved with the project on daily basis but also
with the actual modelling effort itself.

MDBS Case 8

Uploaded by

Copyright:

Available Formats

MDBS Case 8

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MDBS Case 8

Uploaded by

Copyright:

Available Formats

Case 8: Data warehousing and multi-dimensional data modelling

Introduction and case facts

Criteria for evaluation

1) Come up with a visual architecture of the data warehouse

2) Model the details just in time (JIT)

3) Prove the architecture early

You might also like