Manufacturing

White Paper

Realizing a Demand Driven Supply Chain

About the Authors

Lakshmi Narasimhan B.S

Lakshmi Narasimhan is a Senior Business Consultant with the Innovation and
Transformation Group – the cutting-edge domain consulting group within the
Manufacturing business unit at Tata Consultancy Services (TCS). He has over 15 years of
experience spanning the areas of supply chain consulting and IT. He has implemented
solutions based on the Theory of Constraints (ToC) method across production and
distribution operations, and project management. He is a certified practitioner of ToC
and Demand Driven MRP (DDMRP) methodologies.

Ramesh Srinivasan
Ramesh Srinivasan is a Consulting Partner with the Innovation and Transformation
Group in the Manufacturing business unit at TCS. He has over 25 years of experience
spanning the entire value chain, from product development, manufacturing, and supply
chain to product costing and marketing. He has led several supply chain improvement
initiatives for consumer electronics, high-tech, aerospace, and industrial clients,
including the development of a unique outsourced International Purchasing Office.
Abstract

Ever increasing product variety and a rapidly shortening product lifecycle make it difficult for
companies to predict what their customers may buy and when. In addition, the global nature
of supply and demand increases the complexity of supply chains. To remain profitable and
stay relevant to customers in this environment, businesses need to maintain high customer
service levels, optimal working capital, and tight controls over cost. Most companies have
adopted formal planning methods such as Material Requirements Planning (MRP)
implemented through a variety of information technology solutions including Enterprise
Resource Planning (ERP) and Advanced Planning and Scheduling (APS) systems. However, the
right balance between working capital and service levels continues to elude many companies.

One of the key reasons is the lack of a robust supply chain management strategy that
effectively addresses the needs of today's business landscape. For instance, deterministic
formal planning methods like MRP are still in use, despite not being able to cope with today's
level of variation and uncertainty.

Over time, various improvement methods such as Lean, Six Sigma, Theory of Constraints
(TOC), and Total Quality Management (TQM) have been developed to overcome the
challenges in supply chain management. These methods mainly focus on making businesses
more customer centric or demand (market) driven. Several companies have adopted these
methods, and have enjoyed some degree of success. However, most of them did not embrace
an end-to-end demand driven strategy due to various misconceptions around these
improvement methods, including the fundamental distinction between 'push' and 'pull'
strategies. As a result, these methods failed to deliver the desired results.

In this paper, we discuss the need to make over MRP in the context of today's business
requirements as well as the misconceptions surrounding the improvement methods. It also
highlights an innovative 'position and pull' method, which is an advanced consumption
based planning method that enables companies to become truly demand or market driven.
Finally, the paper outlines the key steps to implement this method.
Contents

Surviving the New Supply Chain Reality 5

Material Requirements Planning: Too Rigid for Agility 5
Evolution of Planning and Execution Improvement Methods 6
Pull-based Systems: Looking Beyond Misconceptions 7
Misconception #1: Pull is Make-to-Order 7
Misconception #2: Customer Order Decoupling Point (CODP)
is the push-pull boundary 8
Misconception #3: Pull systems do not have anything to do with forecasting 9
Misconception #4: Pull is all about making suppliers stock up 9
Misconception #5: Pull is only for raw materials and low value components 9
Adopting a Demand Driven Approach through Strategic Decoupling 10
Focusing on near-term demand using pull method 10
Role of forecasts in pull-based systems 10
Three Steps to Transforming to a Demand Driven Supply Chain 11
Step 1: Identify strategic decoupling points for inventory positioning 11
Step 2: Implement effective pull solution 11
Step 3: Drive efficient change management 11
Getting the Technology Right 12
Using the DDMRP framework: A case study 12
Building a Responsive Supply Chain 13
Surviving the New Supply Chain Reality
With the proliferation of product varieties, and distribution channels and locations, supply chain management has
become more complex than ever before. While prior to the 1980s, the markets were primarily supply driven, today
manufacturing companies need to respond effectively and swiftly to changing demand patterns. In other words,
supply chains need to be demand driven to help manufacturers maintain high customer service levels with optimal
working capital. This in turn drives profitability in today's highly variable, uncertain, and complex business
environment.
However, the shift to a demand driven mode is made difficult by poor predictability of demand. There is no definite
way to foretell which products will sell, in what quantity, where, and at what price. In addition, intense competition
and eroding profit margins are forcing companies to reduce costs by outsourcing manufacturing to low-cost
countries, which in turn, increases the complexity of supply chains.
Companies thus adopt several supply chain initiatives such as improving supply chain visibility, implementing ERP
and APS, vendor managed inventory, and so on. However, these initiatives fall short of driving lasting
improvements as they often do not deal with the root cause of the issue. This makes it critical for supply chain
managers and business leaders to understand the fundamental issues at play and invest in appropriate solutions.
We address the two key aspects which prevent companies from realizing a truly demand driven supply chain:
1. The use of an outdated formal planning system like MRP, which is inappropriate for today's highly variable and
uncertain business landscape
2. Misconceptions around the pull-based system and the resulting failure to adapt it to today's business landscape

Material Requirements Planning: Too Rigid for Agility

Most manufacturing and distribution companies have implemented some form of an ERP system to support their
planning and execution processes. However, most ERP systems employ the Material Requirements Planning (MRP)
method conceived in the early 1960s as the backbone for formal planning. Essentially, MRP uses deterministic logic
to calculate component requirements (based on the Bill of Materials – BoM) and schedules the production or
procurement of components based on the sales forecasts or confirmed customer orders for finished goods. The
challenge here is that when the input forecast is inaccurate, which is often the case, all of the derived component
quantities and schedules are also inaccurate.
Furthermore, variability in one part of the system is amplified across the system, impacting the overall performance.
For example, any small change in the plan such as unavailability of a machine or a component has a cascading
impact on the overall plan. Hence, it is difficult for planners to use the output of MRP for executing day-to-day shop
floor activities. In addition, companies use economic order quantities (EOQ) to optimize transportation or
production costs, further distorting the demand signal. Material requirement planners therefore use spreadsheets
for day-to-day execution planning, despite investing in state-of-the-art ERP or APS systems.

5
While several efforts have been made over the years to improve the formal planning model, the basic framework of
MRP within most of the popular ERP systems remains the same as it was originally conceived in the early 1960s. The
unsuitability of traditional MRP systems for today's business landscape leads to misaligned inventories, high costs
associated with expediting production, and the inability to ensure on-time delivery. These factors, in turn,
negatively affect sales and increase the working capital requirement.
A leading industrial goods manufacturing company based in Europe, for example, relied on forecast-based MRP for
production and inventory planning, and used spreadsheets for daily execution. ERP was limited to automating
procure-to-pay process, and the business environment was characterized by intermittent demand, and high
variability in demand and supply.
The company's business unit had an inventory turnover of about 4, while on-time delivery (OTD) based on
customer requested date was about 70%. The root cause analysis of poor OTD and low inventory turnover,
depicted in Figure 1, showed that about 50% of inventory was in excess, of which 35% could directly or indirectly
be attributed to inaccurate forecasts. This clearly proves the damaging consequences of using inappropriate
methods to manage today's supply chains.

Causes for Non-Optimal Inventory

Customer Related Lack of effective collaboration with customers 11%

Excess and obsolete inventory (non-moving stock) due

20%
to inaccurate forecasts 35% of Excess
Internal Process Related Inventory due 50% Excess
Misalignment between inventory and demand (in cycle
8% to forecast Inventory
stock) due to inaccurateforecasts
inaccuracy
Supplier volumeagreements based on forecasts 7%
Supplier Related
Lot sizes greaterthan optimal 4%

Total inventoryanalyzed 100%

Figure 1. Root cause analysis of OTD and inventory turnover

Evolution of planning and execution improvement methods

Demand variability and uncertainty have been longstanding challenges for most business operations. This is why,
since the 1970s, we have seen the evolution of various methods to help improve business planning and execution
processes.
n Lean (1970s): Focuses on reducing waste to expose hidden problems such as poor processes, with the goal of
fixing them in order to help manage uncertainty and increase system reliability
n Theory of Constraints (1980s): Highlights the need to focus on system constraints in order to manage variation
and uncertainty better
n Six Sigma (1980s): Aims to improve quality in order to reduce variation
n Factory Physics (1990s): Uses inventory, capacity, and time buffers effectively to manage variability
6
Each of these improvement methods aims to handle variation and uncertainty better through various techniques,
instead of trying to predict the future or buffer the system through excess inventory.
A key technique that emerged out of the Lean methodology is the 'pull' method that was subsequently improved
upon and is today known as Demand Driven MRP (DDMRP)¹. DDMRP aims to strategically 'position' inventory
buffers in both the BoM and the distribution network to dampen the impact of variability. In effect, the pull method
supplemented by effective inventory positioning offers manufacturers a viable method to implement an
enterprise-wide demand driven supply chain.

Pull-based Systems: Looking Beyond Misconceptions

Although the discussed improvement methods help address the challenges posed by the changing business
landscape, they have not seen wide adoption. Some of the reasons for this include misconceptions surrounding
the pull-based systems and the deep cultural change required for successful implementation. In addition, the fact
that ERP packages did not evolve to support these improvement methods compounded the challenge.
According to Factory Physics, the definition of a pull system is as follows:²
Definition (Pull and Push): A pull production system is one that explicitly limits the amount of work in process that can be
in the system. By default, this implies that a push production system is one that has no explicit limit on the amount of
work in process that can be in the system.
In a pull system, whenever the inventory level goes below the established maximum limit (target level), new orders
are released into the system, but only up to the target level. The target level is based on estimated demand during
the replenishment lead time factored in for variability. Based on the changes to the demand pattern, the target
level can be changed from time to time.
Let's take a look at some of the main misconceptions that make the adoption of the pull system challenging.

Misconception #1: Pull is Make-to-Order

According to popular notions, pull is customer order driven, while push is based on forecasting. In other words pull is
Make-To-Order (MTO), while push is Make-To-Stock (MTS). Though at a strategic level this distinction makes sense, at
an operational level it does not. This can be understood in the context of a supermarket. The supermarket cannot be
customer order driven in an operational sense as it cannot start procurement after the customer walks into the store.
This implies that the supermarket needs to stock its goods before the customer arrives. Hence it must operate on an
MTS mode based on some form of forecasting. Can it be categorized as a 'push' system because of its dependence
on forecasting? No. The idea of 'pull' originated when Ohno, who conceived the Toyota Production System (TPS),
observed replenishment of the supermarket shelf from its storeroom, wherein the supermarket replenishes only
what has been picked up by the customers³. Pull systems apply the same principle to the factory floors.

[1] Orlicky’s Material Requirements Planning, Smith, C., and Ptak, C., McGraw-Hill Professional, 3rd Ed (2011)
[2] Factory Physics for Managers: How Leaders Improve Performance in a Post-Lean Six Sigma World, Mark L. Spearman, Edward S. Pound and Jeffrey H. Bellund,
McGraw-Hill (2014)
7
[3] The Economist, Taiichi Ohno, July 3, 2009, accessed March 2016, http://www.economist.com/node/13941150
Thus, while the push system is entirely based on forecasts, the pull system involves placing orders upstream in the
supply chain. These orders are triggered when actual customer demand, or usage by a production process,
depletes inventory to a point that compromises the ability to serve estimated future demand during replenishment
lead time.

Misconception #2: The Customer Order Decoupling Point is the push-pull

boundary
The supply chain can be depicted, at a high level, as having two parts (MTO and MTS) separated by an inventory
buffer point known as the Customer Order Decoupling Point (CODP), as shown in Figure 2. Depending on where
the CODP is located, the fulfillment strategy is classified as MTS, Assemble-to-Order (ATO), or MTO.
It is often erroneously assumed that the part after the CODP is always pull and the part preceding the CODP is
always push. Therefore, the CODP is often considered to be the push-pull boundary. In reality, CODP can be either
push based or pull based, depending on how the CODP or the inventory buffer point is planned and replenished. If
it is based on forecasts, then it would be right to say that it is a push system, but if it is replenished based on a pull
trigger, then it is actually a pull system.
To reduce lead times on customer order fulfillment, companies typically adopt the CODP approach. But in most
cases, the CODP point is planned based on forecasts, leaving the efficacy of the supply chain to the mercy of
forecasting accuracy. Companies can significantly improve their response to demand signals by planning CODP
based on pull triggers. In addition to CODP, there can be other additional (strategic) decoupling points that help the
supply chain become truly demand driven.

CODP

Make-to-Stock MTS

CODP

Assemble-to-Order MTS MTO

CODP

Make-to-Order MTO

CODP

Push? Pull?

Figure 2 : Positioning of CODP in the supply chain

8
Misconception #3: Pull systems do not have anything to do with forecasting
It is erroneously believed that pull systems do not leverage forecasting. In a pull system, unlike traditional
forecasting based MRP, forecasts need not be broken down into definite time buckets or time series. Pull systems
are based on an established inventory level (target level) that triggers the replenishment order, whenever the
actual inventory level falls below the target level.
This target level is based on an estimate of likely demand within a given replenishment lead time. Hence, even the
pull system includes an element of forecasting or estimation of future demand. The key difference, however, is in
the way the forecast information is used. Hence, the pull system need not be solely consumption based, and can be
set up to use any information about future demand. It can also be set up to plan for seasonality and take into
account known order spikes (larger than normal firm demand). But this approach to pull is rarely adopted mainly
because companies wrongly think that it will adversely affect their inventory levels.

Misconception #4: Pull is all about making suppliers stock up

Another notion is that pull is all about making the companies' suppliers stock components on their behalf, enabling
companies to pull the components from suppliers when required. This concept is used in vendor managed
inventory (VMI). Closely associated with this notion is the erroneous assumption that the replenishment lead time
should be extremely short to implement pull. This has led to the creation of an ecosystem of suppliers with
manufacturing facilities or warehouses in close proximity to their customers, as is common in the automotive
industry. The problem with this notion is that companies become dependent on their suppliers to implement pull.
Furthermore, merely transferring inefficiencies to the suppliers is not a sustainable proposition.

Misconception #5: Pull is only for raw materials and low value components
Many companies implement pull replenishment mechanisms only for C-class (less critical or low value) parts or
parts that experience steady demand, as they do not have to change the re-order points and safety stock levels
often. Due to this, opportunities for holding optimal inventories for A and B class items are missed.
With all of these misconceptions, companies believe that they cannot move to a pull-based system if they cannot
implement MTO, give up on forecasting, or make their suppliers stock up on their behalf. However, by gaining a true
understanding of 'pull' and 'positioning,' (strategic decoupling points) it is possible to implement an end-to-end,
flexible demand driven system that is less deterministic than the traditional MRP.

09
Adopting a Demand Driven Approach through
Strategic Decoupling
The key to managing demand-supply mismatches and improving supply chain performance includes removing the
misconceptions around the pull approach, setting up effective decoupling points (inventory positions) across the
system, and using an effective pull system to manage these inventory positions. These decoupling points act as
shock absorbers and prevent the propagation of variability across the system. Effective positioning of inventory
buffers within the BoM structure as well as the distribution network can deliver greater benefits compared to
conventional tactics such as demand variability reduction, complexity reduction, and forecast accuracy⁴.

Focusing on near-term demand using the pull method

In this approach, pull-based supply order generation is governed by an upper limit or target level in the system. The
target level is based on short-term demand projections that are more accurate than longer-term forecasts. Larger
than normal firm demand is planned additionally, thus preventing it from overwhelming the planned stocks for
normal or regular demand.
This process of generating supply orders offers the following benefits:
n Better ability to manage the current demand situation without building excess inventories, which would be the
result of using inaccurate forecasts
n Flexibility to adjust the target level in tune with the changing demand signals

Role of forecasts in pull-based systems

One of the drawbacks of planning for only short-term demand projections is the inability to deal with situations
that need more time (where the planning horizon is longer than the lead time expected by customers). This
includes building inventories for seasonal products, new product introductions, and large capacity adjustments.
This also applies in cases where suppliers or upstream resources have practical limits on how much they can supply
even if the order is placed ahead of lead time.
Companies can overcome this drawback by using longer term forecasts to adjust the inventory target level to
manage known seasonal events, upsize or downsize production capacities, as well as the launch or phasing out of
products.

10
[4] Opsrules.com, “Everything you ever wanted to know about inventory”, 2014, http://www.opsrules.com/white-papers, accessed December 2, 2015
Three Steps to Transforming to a Demand Driven
Supply Chain
Aligning processes, people, technology, and measurements is an important element in a transformation journey.
We recommend that organizations embed the following three steps in their transformation to a demand driven
supply chain:

Step 1: Identify strategic decoupling points for inventory positioning

The main focus here is to identify nodes in the BoM or locations in the supply chain network that can act as
strategic decoupling points. The key criteria to identifying these points include customer tolerance time, market
potential lead time (or a shorter lead time for which the customer is willing to pay a premium), supply and demand
variability, BoM, supply and distribution network, and critical resource considerations. Positioning of decoupling
points where demand variability is lower (for example at a central warehouse) helps improve customer service, and
eliminates the need to hold inventories at every retail location.

Step 2: Implement an effective pull solution

Each decoupling point, or inventory buffer, should be replenished based on the pull mechanism through the
following measures:
n Estimate the inventory or target level based on the traits of each part such as variability and lead time. Use
simulation and what-if scenarios to arrive at effective values.
n Set up a pull mechanism to replenish the decoupling point.
n Adjust the target level dynamically, in tune with demand patterns and the company's ability to invest up to the
target levels.
n Use long-term forecasts to set the target level for seasonal events and capacity changes.
n Consider any sporadic large orders (beyond a threshold) as additional requirements to ensure that they do not
overwhelm the inventory buffers or result in longer lead times for large orders.
n Drive visible and collaborative execution based on stock buffer priority rather than due date priority.

Step 3: Drive efficient change management

Planners and other upstream and downstream stakeholders must be made aware of the misconceptions about pull
and push, and the positive impact of practicing the new method on supply chain performance. Companies can
initially conduct general training and game sessions to remove misconceptions, followed by collaborative
workshops conducted at regular intervals to help the organization shift to a demand driven supply chain.

11
According to Gartner, transformation into a demand driven supply chain requires a system of technologies and
processes that senses, shapes and responds to real time demand signals across a supply network of customers,
suppliers and employees⁵. However, even today, responding to real-time demand signals is a chronic pain area for
most organizations. The recommended three steps can be an effective approach to solve this problem.

Getting the Technology Right

A majority of the current breed of IT solutions fail to support companies in realizing a demand driven supply chain.
Popular ERP packages are based on outdated and inadequate MRP frameworks, and even the newly developed
modules and features do not address this fundamental problem.
In order to ensure a smooth transition and deliver superior results, companies need to either:
n Choose an appropriate add-on ERP software system with the required solution characteristics, or
n Customize the existing ERP system to support the described position-and-pull processes
Both options have pros and cons depending on factors such as the ERP packages used, and the nature and culture
of the organization. Companies need to decide what's best for them based on a comprehensive analysis of the
situation.

Using the DDMRP framework: A case study

Let's see how a large company analyzed its inventory levels leveraging the DDMRP framework to estimate the
potential for improvement.
A leading power and automation technology company conducted an inventory analysis in order to:
n Release cash locked in working capital as inventory
n Drive inventory re-alignment and optimization for better service levels coupled with lower inventory levels
The company used the Demand Driven MRP framework to calculate the 'should be' inventory for a total closing
stock worth around USD 250 million and compared it with the company's current inventory levels. The results of the
study showed that inventory could potentially be reduced by over 21%. But more importantly, the company found
that the framework can help it align its inventories with demand patterns, thus improving fill rates and on-time
delivery. This, in turn, is likely to have a substantial impact on its sales revenue and net profits.

[5] Gartner, “Demand-Driven Value Network Orchestration Key Initiative Overview”, 2012, https://www.gartner.com/doc/1900216/demanddriven-value- 12
network-orchestration-key, accessed April 1, 2016
Building a Responsive Supply Chain
Manufacturers require effective tools and solutions to revamp their supply chain strategies and live up to evolving
customer expectations. This is truer of industries such as high tech that are characterized by shorter product
lifecycles and greater market volatilities. As a result, manufacturing organizations across industries are increasingly
shifting to enterprise-wide demand driven planning, where production and procurement are driven by
consumption-based replenishment based on a 'position and pull' strategy. This approach helps companies increase
the productivity of their supply chains, reduce wastage, and optimize working capital.
To achieve an enterprise-wide demand driven supply chain, manufacturers will need to bring about a cultural shift
in their organizations. The misconceptions discussed in this paper are at the heart of many partial and failed
adoptions of the pull method. Addressing the reasons for these misconceptions, re-educating supply chain
stakeholders, and winning their buy-in will be key to successfully transitioning to a demand driven system.
Finally, by deploying innovative technologies to implement the position-and-pull strategy, manufacturers can
effectively tackle the challenging business landscape of today to achieve faster growth and greater productivity.

13
About TCS' Manufacturing Business Unit

TCS helps global manufacturers reduce operational expenditure, utilize capacity optimally, and
increase efficiencies while meeting safety and regulatory norms. We are the preferred partner for a
third of the Fortune 500 manufacturers, and have a record of enabling business innovation that
helps them meet the objectives of global operations.

The core strength of our solutions lies in our rich experience across discrete (automotive, industrial
manufacturing, and aerospace) and process industries (chemicals, cement, glass, and paper). Our
vertical focused Centers of Excellence (CoE) leverage this rich database to cross-reference learning
and drive innovation in business solutions for standardized processes, assets and templates, ERP
implementation, and continued support services.

Our solutions and services portfolio spans IT-led business transformation; design, development, and
support for IT solutions; and value-added services such as infrastructure management and
consulting.

Contact
For more information about TCS’ Manufacturing Business Unit, visit:
http://www.tcs.com/industries/manufacturing/Pages/default.aspx
Email: manufacturing.solutions@tcs.com

Subscribe to TCS White Papers

TCS.com RSS: http://www.tcs.com/rss_feeds/Pages/feed.aspx?f=w
Feedburner: http://feeds2.feedburner.com/tcswhitepapers

About Tata Consultancy Services (TCS)

Tata Consultancy Services is an IT services, consulting and business solutions organization that
delivers real results to global business, ensuring a level of certainty no other firm can match.
TCS offers a consulting-led, integrated portfolio of IT and IT-enabled infrastructure, engineering and
assurance services. This is delivered through its unique Global Network Delivery ModelTM,
recognized as the benchmark of excellence in software development. A part of the Tata Group,
India’s largest industrial conglomerate, TCS has a global footprint and is listed on the National Stock
Exchange and Bombay Stock Exchange in India.

For more information, visit us at www.tcs.com

IT Services
TCS Design Services I M I 05 I 16

Business Solutions
Consulting

All content / information present here is the exclusive property of Tata Consultancy Services Limited (TCS). The content / information contained here is correct at
the time of publishing. No material from here may be copied, modified, reproduced, republished, uploaded, transmitted, posted or distributed in any form
without prior written permission from TCS. Unauthorized use of the content / information appearing here may violate copyright, trademark and other applicable
laws, and could result in criminal or civil penalties. Copyright © 2016 Tata Consultancy Services Limited