A DESIGN FOR BUSINESS INTELLIGENCE SERVICE IN
DEMAND DRIVEN SUPPLY CHAIN MANAGEMENT
T. He
, P. Ribbins
, and R. Brown
,
†Orchestr8 Ltd., United Kingdom
L. Sun
and C. Gurr
‡School of Systems Engineering, University of Reading, United Kingdom
Keywords: Demand driven supply network, business intelligence services, metrics reporting service, and supply chain
management processes.
Abstract: This paper discusses the problems inherent within traditional supply chain management’s forecast and
inventory management processes arising when tackling demand driven supply chain. A demand driven
supply chain management architecture developed by Orchestr8 Ltd., U.K. is described to demonstrate its
advantages over traditional supply chain management. Within this architecture, a metrics reporting system is
designed by adopting business intelligence technology that supports users for decision making and planning
supply activities over supply chain health.
1 INTRODUCTION
Supply chain management has been evolving since
the 1980s through various stages, such as cost
realignment, business process reengineering, and
vertical integration in the value chain. Recently there
has been an increasing concern regarding an
emergence of demand-driven supply chain
management methodology. Traditional supply chain
management improvement approaches in the
manufacturing industry are based mostly on
improving capacity and internal efficiency and
attempting to increase forecasting accuracy to
support better inventory planning using MRP
systems. However, to deal with the challenges of
managing today’s supply chains, characterised by
ever decreasing lead time demands from customers
allied to a need for greater choice and customisation,
responsiveness and agility have to be considered as
part of the main principle.
Adjusting to this change in supply chain
management, certain supply chain processes need to
be redesigned, e.g., sales and operations planning,
pull replenishment logic, and timely demand capture.
The U.K. based supply chain management
consulting company, Orchestr8, has developed a
demand-driven supply chain management
methodology which is now being implemented with
a number of its blue chip multi-national clients. The
development of the demand driven supply chain
management system with these major clients has
produced a range of business and technical
components which advance the management of the
supply chain from supply-driven to demand-driven.
The software system has benefited its clients to align
its daily supply chain operations with long-term
supply chain strategies which in return reduces
inventory cost and builds up stable relationship with
their suppliers. This paper describes the architecture
of Orchestr8’s demand-driven supply chain
management system and its conceptual design
toward a business intelligence based metrics
reporting service.
2 SUPPLY CHAIN
MANAGEMENT PROCESSES
Supply chain is featured by Lambert et al. (Lambert
et al , 1998) as a supply network which coordinates
organisations, people, activities, information and
resources involved in moving a product or service
20
He T., Ribbins P., Brown R., Sun L. and Gurr C. (2007).
A DESIGN FOR BUSINESS INTELLIGENCE SERVICE IN DEMAND DRIVEN SUPPLY CHAIN MANAGEMENT.
In Proceedings of the Ninth International Conference on Enterprise Information Systems - SAIC, pages 20-25
DOI: 10.5220/0002412700200025
Copyright
c
SciTePress
from supplier to customer. Judging the effectiveness
of supply chain management normally includes
monitoring delivery and order fulfillment
performance, production flexibility, warranty and
return’s processing costs, inventory and asset turns,
and other factors in evaluating the overall
performance of a supply chain (Poluha, 2006). To
achieve excellent supply chain management, it is
important to understand the supply network and its
detailed processes.
Figure 1: Structure of Supply Network (Lambert et al.,
1998).
A supply network as shown in Figure 1 consists
of a focal entity of a manufacturing company
networked by downstream entities of customers and
upstream entities of suppliers. In the current
information era, the supply network becomes more
complex and challenging when dealing with the
dynamics of upstream suppliers and highly uncertain
international markets’ demands. Therefore, the
supply chain management is seen as an information-
intensive process, where material/products delivery
and demands have to be effectively managed in
order to maintain supply chain’s visibility.
Because of the large amount of business data in
supply chain management, most companies
implement enterprise information systems such as
ERP, MRP or supply chain management modules to
improve supply chain workflow and efficiency.
However, these systems have not provided sufficient
impact on the effectiveness in the workflow, as two
influential functions, i.e., forecasting and inventory
management, play limited roles in the supply
network.
Forecasting concerns the issues such as how to
project historical data; how to deal with the change
of lead time; and how to recognise different demand
patterns regarding various components (Donald,
2003). Warwick Manufacturing Group (WMG, 2003)
points out that the critical reasons for failing in
getting quality forecasts are:
Fails to have a single formal forecast
process by all of the management and all the
parties;
Fails to monitor forecast accuracy
over different time periods and constantly
review the forecast as it changes from time to
time.
More often than not, in manufacturing business,
an ineffective forecast can cause a production plan
ending up with either pushing too much production
and excessive stock, or totally running out of
products.
Inventory is summarised by Richards (Richards,
1982) that time, discontinuity, uncertainty, and
economy feature in the existence of inventory.
Inventory is expected to work as a buffer between
supply and demand to minimise the risk of
investment. Therefore, there are problems
concerning what to stock, when to stock and how to
stock. It is no doubt that a holistic view over parts’
behaviour and aligning that with an appropriate
supply chain management strategy could take
inventory management into a more strategic level
and thereby enhance the effectiveness in setting
stock policies. However, some companies find that
this process, normally called inventory or stock
management, is complex in terms of the diversity of
parts and the time involved.
These limitations of the current common
approach to supply chain management may hinder a
company’s ability to achieve its business goals.
Consequently, companies may find themselves
losing their competitive advantages. Inspired by its
various experiences in tackling supply chain
management difficulties and issues, Orchestr8
realises that the traditional supply network needs to
be driven by real-time demand.
3 DEMAND DRIVEN SUPPLY
CHAIN MANAGEMENT
ARCHITECTURE
Demand driven supply network (DDSN) is a system
of coordinated technologies and processes that
senses and reacts to real-time demand signals across
a network of customers, suppliers, and employees. It
enables organisational efficiency, streamline new
product development and launch, and maximise
A DESIGN FOR BUSINESS INTELLIGENCE SERVICE IN DEMAND DRIVEN SUPPLY CHAIN MANAGEMENT
21
margin (Askegar, 2004). The main characteristics of
DDSN are featured as follows (Laura, Kevin and et
al, 2004).
Business manufactures to demand, not
inventory or capacity
;
Processes are to be integrated across the supply
chain
.
Orchestr8 adapted this approach and developed a
demand driven supply chain management
architecture (see Figure 2). One of its clients, who is
a market leading manufacturer and retailer focusing
on healthcare products with turnovers in excess of
£5 billion, has benefited from the demand driven
supply chain services. This client is based in the
United Kingdom and having branches in Australia,
Canada and many other countries; and its supply
network extends from European countries to East
Asia. As a typical large scale international
manufacturing and retailing company, the client
found it becoming increasing difficult to maintain its
competitive advantages in the value chain. High
percentages of excess stocks, low efficiency in
managing supplier’s delivery and not having the
right stock at the right time in the right place
jeopardise the manufacturing effectiveness.
Orchestr8 took on these problems of this client and
provided the solution based on a demand driven
supply chain management process.
Figure 2: Orchestr8’s DDSC Management Architecture.
Supported by Orchestr8’s three tier service
model, its demand driven supply chain management
process assimilates the data integration, Sales and
Operations Planning, replenishment planning and
supply chain integration. Orchestr8’s business
intelligence based metrics reporting system
facilitates their forecasting and real-time inventory
monitoring within the supply network.
3.1 Service Model
The service model (Figure 3) in the architecture
provides a supply chain management infrastructure
on which stock turnover, supply and demand
stability and other metrics can be measured. In order
to establish a holistic view on the supply network, a
planning for demand driven supply should be carried
out at the strategic, tactical, and operational levels.
Sales & Operations Planning (S&OP) enables
Orchestr8’s clients and their suppliers to build a
mutual understanding of the inventory strategies
selected for all the stock items.
Figure 3: Orchestr8 Service Model (Orchestr8 Ltd., 2005).
Important changes in product life cycle or market
demand will also be drawn for attention to reach a
better demand visibility. Based on the inventory
strategies formulated by the S&OP process, daily
replenishment is then planned at the operational
level that triggers orders when necessary. A web-
based metrics system monitors the supply activities
by producing timely analysis in reports for decision
making processes.
3.2 Process Model
The process model (Figure 4) is designed to
streamline the inventory management process.
Figure 4: Orchestr8 Process Model (Orchestr8 Ltd., 2005).
ICEIS 2007 - International Conference on Enterprise Information Systems
22
From the diagram, it can be seen that the monthly
process is the one where the inventory strategies are
defined for the next period. These strategies are
created based on the intelligence gathered as part of
the S&OP process. An automated replenishment
order-generating process then ensures that the
inventory is managed in accordance with the policies.
3.3 Sales and Operations Planning
The importance of the S&OP process arises from its
ability in gathering all the stakeholders’
requirements to reach a common understanding of
the most effective inventory strategies. The S&OP
process also provides the opportunity of realigning
the clients’ forecast by ensuring that the ‘plan’ is
conditioned to incorporate other influences such as
seasonality or promotions management. Orchestr8’s
forecast models allow these options to be tested
through various what-if simulations which can be
done at various levels e.g. by individual item, by
product family or by common grouping.
3.4 Demand-Pull Replenishment
Planning
The replenishment planning process incorporates
inventory items’ historical demand patterns with
their current replenishment rules. Figure 5 shows the
analysis of parts regarding their daily usage and
demand variability based on historical archives.
Figure 5: Volume & Variability Analysis (Orchestr8 Ltd.,
2005).
The rules can be applied by product family or
individual item to achieve most effective
replenishment planning. Vendor Managed Inventory
and Kanban replenishment logic are deployed where
appropriate to enhance lean manufacturing which
results in benefits such as lead time improvement
and stock reduction.
3.5 Business Data Integration
Business data, including order information and part
master, is essential to the successes of effective
forecasting and inventory management. Orchestr8’s
system can be integrated with most ERP, MRP and
other workflow systems and carries out a validation
function on the data imported to ensure that the
necessary data is in complete and accurate which in
turn improves the consistency within the supply
chain management processes. During the restrict
data validation process, the system filters corrupted
data out with relevant implication about its problems
and meanwhile, allows correct data to pass through
for further planning which dramatically reduces
potential problems in the supply chain management
process.
3.6 Supply Chain Integration
Table 1: Traditional SCM vs. Demand-driven SCM.
Topic
Traditional SCM
v.s.
Demand-driven SCM
Scope of
Management
Focus on
internal process
Focus on supply
chain as a whole
Forecast
Effectiveness
Inaccurate and
not understood
by all of the
management
Using demand
driven forecast
model & agreed
by all of
management
Demand
Capture
Push and
produce to
production plan
method
Pull and produce
to demand
Inventory
Management
Efficiency but
not managed as
per various parts’
behaviour
Effective and
detailed to each
product family or
individual item
React to
Change
Difficult to
achieve as the
core calculation
is rigid
Flexible based
on improved
supply network
with lean
inventory &
shortened lead
time
Organisational alignment is always a key
concern of demand driven supply chain management.
Orchestr8 encourages the collaboration amongst
stakeholders across the supply networks. Inventory
strategies are clearly formulated and monitored by
all the stakeholders periodically. Great efforts are
made on building good relationships with the
suppliers to ensure the supply chain is aligned to
A DESIGN FOR BUSINESS INTELLIGENCE SERVICE IN DEMAND DRIVEN SUPPLY CHAIN MANAGEMENT
23
their capabilities and customers’ expectations.
The demand driven supply chain management
architecture focuses on improving supply chain
management effectiveness in an extensive sense
across the value chain. Table 1 summarises its
business value in comparison with the non-demand
driven approach in supply chain management.
4 CONCEPTUAL DESIGN FOR
BUSINESS INTELLIGENCE
METRICS REPORTING
SERVICES
Business Intelligence (BI) Metrics Reporting Service
plays a strategic role in Orchestr8’s demand driven
supply chain management architecture. It produces
reports and dashboards which support the
collaboration between Orchestr8’s clients and their
suppliers in a real-time supply network. The design
of the metrics reporting system has adopted an
advanced business intelligence approach based on
Kimball’s model (Mundy, 2006) to ensure that this
system has agility.
Various performance measures’ mechanisms
from business intelligence methods have been
critically assessed by this development. The
technologies, such as data warehousing, Online
Analytical Process (OLAP) (Sandeep and Sourabh,
2006), and Microsoft integration and reporting
services have been adopted in the design of the
metrics reporting system (see Figure 6).
Figure 6: Data Transformation (Orchestr8 Ltd., 2006).
The operational data of the clients and its
suppliers is transformed from the Online
Transactional Process (OLTP) database into a data
warehouse which schematically structures the
information for report generation. The cubes
facilitate the execution of the reporting process
which takes the users request on visualising their
supply chain performance measures.
The operational data of the clients and its
suppliers is transformed from the OLTP database
into a data warehouse which schematically
structures the information for report generation. The
BI capable cubes facilitate the execution of the
reporting process which takes the user’s request on
visualising their supply chain performance measures.
Figure 7: The Metrics Reporting Services’ Structure.
A schema of the metrics reporting services,
shown in Figure 7, represents S&OP, supply,
demand and inventory. The S&OP is mainly
represented by forecast and target in its granularity.
The supply, inventory, and demand cover the
relevant information in their functionalities in the
supply network.
Similar to Debra’s method of organising supply
chain metrics hierarchically, the metrics provided by
the reporting service are designed to demonstrate
different degrees of detail at various business levels.
In other words, the structure of the metrics
represents details of the supply chain management
process at its granularity. Figure 8 illustrates a
process of evaluation for a client’s total inventory on
hand performance (Orchestr8, 2006).
This process produces the inventory turns at the
upper level, total inventory value at the middle level,
and the inventory policies by part at the lower level
in the metrics. The inventory turns illustrates the top
level effectiveness of the stock management process
by looking at the ratio between the value of total
usage of the stock and the average value of all the
stock on hand in a certain time period. Indicated by
this measure, managers then can use the inventory
on hand reports which are categorised by suppliers
or by product families or by inventory policies at
their preference to investigate the efficiency of their
inventory strategies. To help clients to obtain further
insight into an individual item or exceptional items,
some special analysis driven reports are available to
cater various decision-making needs.
ICEIS 2007 - International Conference on Enterprise Information Systems
24
Figure 8: A breakdown structure in Inventory Management
performance evaluation.
All in all, the information presented in the
metrics reporting service can support the managers
to assess the overall supply chain health, diagnose
problems, and plan actions progressively.
5 CONCLUSION AND FUTURE
WORK
With respect to the traditional supply chain
management methodology, Orchestr8 in
collaboration with its academic partner has
developed and deployed the demand driven supply
chain management methodology. The development
has delivered the demand driven supply chain
management architecture with the software tools to
perform real-time forecast modelling, inventory
monitoring and supply planning. Clients of
Orchestr8 have benefited by managing their supply
chain on demand. This development has also
provided the client with strategic impact on
redesigning their supply chain planning process.
There are still some further work and
improvements need to be carried out, such as
deploying business intelligence using advanced
analyst methods, incorporating ad-hoc reports in the
metrics reporting services and designing formal
performance management processes. Effectiveness
of the demand driven supply chain management
process will be under close monitor as well for
continuous improvement purpose.
REFERENCES
Askegar, V., 2004. The Demand-driven Supply Network
[Electronic version]. Supply Chain Management
Review, April, 8(3): 15-16.
Cecere, L., O'Marah K. and Preslan L, 2004. Driven By
Demand, [Electronic version]. Supply Chain
Management Review, 8(8): 15-16.
Hofman, D.H., 2004. The Hierarchy of Supply Chain
Metrics, [Electronic version]. Supplier Chain
Management Review, pp: 28-37.
Sandeep, K. and Sourabh D. 2005. Business Intelligence –
Delivering Business Data through Supply Chain
Analytics, [Electronic version]. Infosys Viewpoint,
January. Retrieved January 30, 2007, from
http://www.infosys.com/supplychain/Supply_Chain_A
nalytics.pdf.
Mundy, J. and Thornthwaite W., 2005. The Microsoft Data
Warehouse Toolkit. Wiley Publishing Inc., Indianapolis,
Indiana.
Lamber, D.M., Cooper M.C. and Pagh J.D., 1998. Supply
Chain Management Issues & Research Opportunities,
[Electronic version]. Int. Journal of Logistics
Management, 9(2): 1-19.
Orchetr8, 2005. Investigation on Business Methodology.
Business Process Documentation, Orchestr8 Ltd.
Orchetr8, 2006. Business Intelligence Metrics Reporting
Service Design and its Architecture. Business Process
Documentation, Orchestr8 Ltd.
Orchetr8, 2006. Sales and Operations Planning Report
Pack, Business Process Documentation, Orchestr8 Ltd.
Poluha, R.G., 2006. Application of the SCOR Model in
Supply Chain Management. Cambria Press,
Youngstown, NY.
Tersine R.J., 1982. Principles of Inventory and Materials
Management. Elsevier Science Publishing Co., Inc..
2nd edition.
Warwick Manufacturing Group, 2003. Introduction to
Forecast, Logistics and Operations Management.
Whitepaper, Warwick Manufacturing Group,
University of Warwick.
Waters, D., 2003. Inventory Control & Management. John
Wiley & Sons Publishing. 2nd. Edition.
A DESIGN FOR BUSINESS INTELLIGENCE SERVICE IN DEMAND DRIVEN SUPPLY CHAIN MANAGEMENT
25