2.

Forecasting and Inventories in SC

Forecasting and its role, characteristics of forecasts, classification, forecasting methods, Integrate demand
planning and forecasting through SC. Planning and managing inventories in SC: Goals, role, effects of cyclic
inventory in SC, effect of lot size, Economic order quantity, Quantity Discount, Strategic change

*Key issues in supply chain management

* The strategic level deals with decisions that have a long-lasting effect on
the firm. This includes decision regarding the number, location, and
capacity of warehouses and manufacturing plants and the flow of material
though the logistics network.

* The tactical level includes decisions that are typically updated anywhere
between once every quarter and every year. These include purchasing and
production decisions, inventory policies, and transportation strategies,
including the frequency with which customers are visited.

* The operational level refers to day-to-day decisions such as scheduling,

lead time questions, routing and truck loading.

Inventory Control: The retailer’s objective is to decide at what point to

recorder a new batch of the product, and how much to order so as to
minimize ordering and holding costs.

Why should the retailer hold inventory in the first place?

What is the impact of the forecasting tool used to predict customer
demand?
Should the retailer order less than, more than, or exactly the demand
forecast?

Supply Contracts: Relationships between supplies and buyers are

established by means of supply contracts that specify pricing and volume
documents, delivery lead times, quality, return, and so forth.
The question, of course is whether supply contracts can also be used to
replace the traditional supply chain strategy with one that optimizes the
entire supply chain performance?

In particular, what is the impact of volume discount and revenue sharing

contracts on supply chain performance?

Distribution strategies: Wal-Mart’s recent success story highlights the

importance of a particular distribution strategy referred to as cross-
docking.

This is a distribution strategy in which the stores are supplied by Central

Warehouses that act as coordinators of the supply process and as
transhipment points for incoming orders from outside vendor, but
warehouse do not keep stock themselves.

How many cross-dock points are necessary?

What are the savings achieved using a cross-docking strategy?

How should a cross-docking strategy be implemented in practice?
Is the cross-docking strategy better than the classical strategy in which
warehouses hold inventory?

Outsourcing and procurement Strategies: Deciding what to make internally

and what to buy from outside sources.

How can a firm identify what manufacturing activities lie in its set of core
competencies and thus should be completed internally and what product
and components should be purchased from outside suppliers, because
these manufacturing activities are not core components?

What are the risks associated with outsourcing and how can these risks be
minimized? When you do outsource, how can you ensure a timely supply of
products?
 Product Design: Effective design plays several critical roles in the supply
chain. Most obviously, certain product designs may increase inventory
holding or transportation costs relative to other designs, while other
designs may facilitate a shorter manufacturing lead times?

Finally, new concepts such as mass customization are increasingly popular.

What role does supply chain management play in the successful
implementation of these concepts?

Information Technology and Decision-Support System: Much of the

current interest in supply chain management is motivated by the
opportunities that appeared due to the abundance of data and the savings
that can be achieved by sophisticated analysis of these data. What data
should be transferred; that is, which data are significant for supply chain
management and which data can safely be ignored?

How should the data be analysed and used? What is impact of the internet?
What is the role of electronic commerce? What infrastructure is required
both internally and between supply chain partners? Information technology
and decision-support systems are both available; can these technologies be
viewed as the main tools used to achieve competitive advantage in the
market? If they can, then what is preventing others from using the same
technology?

Customer Value: Customer value is the measure of a company’s

contribution to its customer, based on the entire range of products,
services, and intangibles that constitute the company’s offerings.
Inventory Management and Risk Pooling

You will understand the following issues:

 How firms cope with huge variability in customer demand?

 What the relationship is between service and inventory levels?

 What impact lead time and lead time variability have on inventory
levels?

 What an effective inventory management policy is?

 How buyers and supplies use supply contracts to improve supply

chain performance?

 What approaches can be used to forecast further demand?

Introduction

Managing inventory in complex supply chain is typically difficult, and may

have a significant impact on the customer service level and supply chain
system wide cost.

Inventory appears in the supply chain in several forms:

 Raw material inventory.

 Work in process (WIP) inventory.

 Finished product inventory.

Each of these needs its own inventory control mechanism.

 Example: General motors (Gm) has one of the largest production and
distribution networks in the world.

In 1984 GM’s. Network: 20,000 supplier plants, 133 parts plants,

31 assembly plants, and 11,000 dealers.
Freight transportation costs: approx. Dollar 4.1 billion with 60 % for
material shipments.

GM has implemented a decision tool: reducing the combined corporate

cost of inventory and transportation. Indeed, by adjusting shipment sizes
(i.e. inventory policy) and routes (i.e. transportation strategy), costs should
be reduced by about 26 percent annually.

Need of Inventory control:

 Unexpected changes in customer demand

 Significant uncertainty in the quantity and quality of the supply,

supplier costs, and delivery times.

 Even if there is no uncertainty in demand of supply, there is a need

to hold inventory due to delivery times.

 Economies of scale offered by transportation companies that

encourage firms to transport and hold large inventories.

 In 1993, Dell Computer’s stock plunged after the company

predicted a loss. Dell acknowledged that the company was sharply
off in its forecast of demand, resulting in inventory write-downs

 In 1994, IBM struggled with shortages in the ThinkPad line due to

ineffective inventory management.

Two important issues in inventory management.

1. Demand forecasting 2. Order quantity calculation
Single warehouse inventory example Factors affecting inventory policy:

 First and foremost is customer demand,

which may be known in advance or may be random.

In the latter case, forecasting tools may be used in situations in

which historical data are available to estimate the average
customer demand, as well as the amount of variability in customer
demand (often measured as the standard deviation).

 Replenishment lead time.

 The number of different products.

 The length of the planning horizon.

 Costs, including order cost and inventory holding cost.

Order cost: product and transportation

 Property taxes and insurance on inventories.

 Maintenance costs.

 Opportunity cost, which represent the return on investment that

one should receive had money been invested in something else
(e.g. the stock market) instead of inventory.

 Service level requirements. where customer demand is uncertain,

it is often impossible to meet customer orders 100 percent of the
time, so management needs to specify to acceptable level service.
The Economic Lot Size Model
The classic economic lot size model, introduced by Ford W. Harries in
1915, is a simple model that illustrates the trade-offs between ordering and
storage costs. The model assumes:
 Demand is constant, at a rate of D items per day.

Order Quantity: fixed at Q items per order;

 A fixed cost (setup cost) K, is incurred every time places an order.

An inventory carrying cost, h, also referred to as a holding cost and

is accrued per unit held in inventory per day that the unit is held.

 The lead time, placement of an order and its receipt, is zero.

Initial inventory is zero. The planning horizon is long (infinite).

 Our goal: obtain optimal order policy that minimizes annual

purchasing and carrying costs while meeting all demand (i.e. without
shortage).
Simplified: a real inventory system. The assumption of a known fixed
demand over a long horizon is clearly unrealistic.
 Replenishment of products very likely takes several days, and the
requirement of a fixed order quantity is restrictive.
Orders be received at the warehouse precisely when the inventory
level drops to zero. i. e zero inventory ordering property.
 To find optimal ordering policy in the economic lot size model,
consider the inventory level as a function of time.
i.e. saw-toothed inventory pattern.
Order Time between two successive replenishment as cycle time.
 Since the fixed cost is charged once per order.
Holding cost as the product of per unit per time period as h;
Average inventory level, Q/2; and the length of cycle, T.

The total inventory cost for cycle length T is

ℎ𝑇𝑄
=K+ -- (f+h cost) -----eqn.1
2

Inventory changes from Q to 0 over time T, Demand D is constant /time,

Then, it is must that Q=DT,

To get average cost /unit time, divide eqn.1 by T,

The Avg.Total cost per unit time is,

𝐾 ℎ𝑇𝑄 𝐾 ℎ 𝑄 𝐾 ℎ 𝑄 𝐾𝐷 ℎ 𝑄
= + = + = + = +
𝑇 2𝑇 𝑇 2 𝑄/𝐷 2 𝑄 2
Optimal policy balances holding cost per unit time with setup cost per unit
time (order cost),

Setup cost (Order cost)/unit time, = (KD)/Q

Holding cost/unit time, = (h Q)/2,

Optimal order Quantity, Q is when order cost and holding cost per unit of
time are equal,

(KD)/Q = (h Q)/2

hQ2 = 2KD

𝟐𝑲𝑫
Q *= √
𝒉
 Managing Inventory in the Supply Chain

Multi-facility supply chain that belongs to a single firm.

The objective: to manage inventory so as to reduce system wide cost;
it is important to consider the interaction of the various facilities and the
impact this interaction has on the inventory policy

Retail distribution system with a single warehouse serving a no. of retailers.

* Inventory decisions: by a single decision maker

* Access to inventory information at: each of retailers and warehouse.

Echelon inventory is an effective way to manage system.

In distribution system, each stage of level (i.e. the warehouse or the

Retailers) often is referred to as an echelon.

 Thus, the echelon inventory at any stage or level of the system is equal
to the inventory on hand at the echelon + downstream inventory.
 For ex. The echelon inventory at the warehouse = inventory at the
warehouse, plus all of inventory in transit to and in stock at retailers.
 Echelon inventory position at the warehouse = inventory at the
warehouse, plus those items ordered by the warehouse that have not
yet arrived minus all items that backordered.
 This suggest the following effective approach to managing the single
warehouse multi retailer system.
 First, the individual retailers are managed as, using the appropriate (s,
S) inventory policy. Second, the warehouse ordering decisions are
based on the echelon inventory position at the warehouse.
3.6 Practical Issues
Effective inventory reduction strategies: The top seven strategies are

1. Periodic inventory review: at a fixed time interval and every time it is

reviewed, a decision is made on the order size.
Review policy makes it possible to identify slow-moving and obsolete
products and allows to continuously reduce inventory levels.

2. Tight management of usage rates, lead times, and safety stock: Such an
inventory control process allows the firm to identify, for ex, situations in
which usage rates decrease for a few months.

3. Reduce safety stock levels: This can perhaps be accomplished by

focusing on lead time reduction.

4. Introduce or enhance cycle counting practice: This process replaces the

annual inventory physical inventory count by a system where part of the
inventory is counted every day, and each item is counted several times
per year.

5. ABC approach: items are classified into three categories.

Class A items include all high-revenue products, which typically account
for about 80 percent of annual (dollar) sales and represent about 20
percent of inventory SKUs.

Class B items: account for about 15 % of annual sales,

Class C products: represent low-revenue items,
Products whose value is no more than 5 % of sales.
 As Class A items: account for the major part of the business,
a high-frequency periodic review policy (e.g., a weekly)
May keeps no inventory of expensive Class….

Class B products: medium frequency of review.

Class C, or keeps a high inventory of inexpensive Class C products.

6. Shift more inventory or inventory ownership to suppliers

7. Quantitative approaches: focus on the right balance between inventory

holding and ordering costs.
Observe that the focus in the survey was not on reducing cost but on
reducing inventory levels Indeed.

Significant effort by industry to increase the inventory turnover ratio,

Inventory turnover ratio = Annual sales

Average inventory level

For instance, retailing powerhouse Wal-Mart has the highest inventory

turnover ratio any discount retailer.

This suggests that Wal-Mart has a higher level of liquidity, smaller risk of
obsolescence, and reduced investment in inventory. Of course, a low
inventory level in itself is not always appropriate since it increase the risk of
lost sales.
Forecasting and its role: Nature of demand
A company must be knowledgeable about numerous factors that are:
• Past demand
• Lead time of product replenishment
• Planned advertising or marketing efforts
• Planned price discounts
• State of the economy
• Actions that competitors have taken
Forecasting components: Approaches, support system, administration,
techniques, errors
Top down and Bottom up approach:
Top Down approach: These are general forecasts which begins with GNP
and NI, for organization as geo political unit from which industry forecast is
developed. Organization share in market is predicted. Then specific
forecasts are developed.

It’s applicable for stable demand situations or demands are changing

uniformly in market.
Forecast is developed for different DCs all together.
 Bottom-up Forecasts: it starts at product level and then aggregate
organizational forecast is made by summing up.
 Which can be modified according to general economic activity, trends
in business, outlook and competitive strength of organization.
 It is decentralised approach and forecast for each distribution centre is
developed separately.
 It can more accurately tracked and consider demand fluctuations
within specific market.
 Requires more detailed record keeping and difficult to accommodate
systematic demand factors, impact of promotion. Correct combination
of detail tracking of bottom-up approach and data manipulation ease of
top down approach.
 In forecast it is very necessary to determine output, its intended uses.
Forecasting methods:
1. Qualitative:
Qualitative forecasting methods are primarily subjective and rely on human
judgment. Most appropriate when little historical data are available or
When experts have market intelligence that may affect the forecast, new industry.
2. Time series:
Use historical demand to make a forecast.
Basic assumption that past demand history is a good indicator of future demand.
Most appropriate when the basic demand pattern does not vary significantly from
one year to next.
These are the simplest methods. Good starting point for a demand forecast.

3. Causal:
Causal forecasting methods assume that the demand forecast is highly correlated
with certain factors in the environment (the state of the economy, interest rates.),

4. Simulation:
Simulation forecasting methods imitate the consumer choices that give rise to
demand to arrive at a forecast.
Using simulation, a firm can combine time-series and causal methods to
answer such questions as: What will be the impact of a price promotion?
What will be the impact of a competitor opening a store nearby?
Airlines simulate customer buying behaviour to forecast demand for higher fare
seats when there are no seats available at the lower fares.
Aggregate Demand Forecasting (function): Economic model building.

 Requires highly technical and professional economists.

Use of mathematical statistics
 A theory to make inference about population (gross no.) on basis of
sample.
 It’s not useful to select representative sample from cross section
study.
 One cannot neglect random variables in relationships which is not
exact.
 So probabilistic thoughts are fundamental consideration in economic
models.
 Forecast may be long-run or short run.

Short run Forecasts:

 Econometric models are not useful for controlling inventories of

individual items which are numerous.
 Forecasting techniques for existing products are different from new
products.
 Existing products: technique of time series analysis of economic
indicators.
 Here, historic data is divided into three elements: demand level,
trend and variation, Seasonal and random.
 Projection of trend by trend line method may be inadequate for short
run method, because of seasonal and cyclic effect.

Different techniques are: Last period demand, Mean average demand,

Moving average method, Correlation and regression analysis.
Moving Average method:
Average of most recent periods.
3, 4 or 12 periods averages are common.
Great amount of historical data is required.

Exponential Smoothing:

 It bases the estimate of future sales on the weighted average of

previous demand and forecast levels.
 The new forecast is function of old forecast incremented by fraction
of differential between old forecast and actual sales realized.
 Rapid calculations of new forecast, without major history of data.
 Increment of adjustment is called alpha factor. Model format:
Example:

Recent time period is 100 days.

Actual sales is 110 units, alpha factor = 0.2, then