ADHYAYAN

TOPICS:

1. Inventory Management

2.Aggregate Production Planning

3.Material Requirement Planning

4.Forecasting
Inventory Management
 Inventory Defined
• An inventory is a stock of an item or idle resource held for future use.
Inventories represent investments designed to assist in
production activities and/or serve customers.
• Inventory is simply a stock of physical assets having some economic
value, which
can be either in the form of material, money or labor.
• Inventory consists of physical items moving through the production
system. The cost of storing inventory accounts for a substantial
portion of manufacturing cost, often 20% or more.
• Different departments within the organization adopt different
attitudes towards
inventory.
 Classification Of Inventory
Manufacturing inventories
• Production Inventory: Items that go into final product also ensure
availability of production such as raw materials, components, sub-
assemblies purchased from outside.
• Work in Process: All items in semi finished form or products at different
stages
of production.
• Finished Product: Final/Completed product ready for
dispatch/shipment to
users/distributors.
• MRO: Maintenance, Repairs and Operating supplies like spares,
consumables which though needed for final product don’t
actually into it – oil, grease, cotton wastes, tools, etc.
 Inventory Mgt. Systems
• ABC
• FSN
• VED
• SDE
• HVL
• XYZ
 EOQ - Assumptions
• Demand for the product is uniform, continuous, constant and deterministic
• The item is replenished in lots or batches, and the quantity need not be an
integral number of
units, and there are no restrictions on its size.
• The unit variable cost does not change appreciably with time;
• The unit variable cost does depend on the replenishment quantity; and there
are not bulk discounts on unit cost or transportation cost.
• The item is treated entirely independently of other items; i.e. benefits from the
joint review or replenishment do not exist or are simply ignored.
• The lead time is zero and the entire ordered quantity is delivered as soon as the
order is placed.
• Delivery is instantaneous, so that all the order arrive at the same time and can
be used
immediately.
• No shortages are allowed. (Shortage cost is prohibitive or very large or infinite)
• The planning horizon is very long and we assume that all parameters will
continue at the same
value for a long time.
 TVC(Q) and EOQ*

TVC(Q)

* * o *
* *

Q
Q*
Optimal Order Size
 BASIC EOQ MODEL
In determining the appropriate order quantity, we use the criterion of minimization of
total
relevant costs; relevant in the sense that they are truly affected by the choice of the
order quantity.
D or d - Demand (or demand rate) of the item. units/unit time.
Q - Order size / replenishment quantity in units
Cu - Purchase price / Production cost - unit variable cost of the item. Rs./unit
Co - Ordering costs / cost of replenishment / setup cost of manufactured items.
The fixed cost component independent of the magnitude of the replenishment
quantity. Rs / order
Ch - Cost of holding the stock. The cost of having one rupee item tied up in the
inventory for a unit time interval usually one year. Rs/unit/unit time. Also
expressed as %charge of purchase price of the item.
t - Cycle time is the time between two consecutive replenishments. This depends
on the order size, with large orders leading to a longer cycle times.
LT - Lead time (delivery lag)
TC (Q) - Total relevant costs Rs. per unit time influenced by the order quantity.
For this model the various levels of stock are as follows:
• Minimum level = safety stock (buffer) = zero
• Maximum level = min level + order quantity = zero + EOQ = Q
• Reorder level = min level + consumption during lead time = zero + LT * D
• Average inventory per cycle = (max level + min level) / 2 = (Q + 0 )/ 2

TC = PC + OC + HC + SC = purchase + ordering + holding + shortage

• PC = price/unit * quantity purchased = Cu * D

• OC = cost of ordering per order * number of orders = Co * D/Q
• HC = cost of carrying one unit * average number of units in stock = Ch * Q/2
• SC = zero (no shortage allowed)

TC = Cu * D + Co * D/Q + Ch * Q/2

• The EOQ is the quantity which minimizes the total costs. Total cost is the sum of fixed cost
and variable cost. Fixed cost component C*D is independent of order size, while the variable
component is dependent on the order size.
Cycle Time:
The cycle time, t, represents the time that elapses between the placement of orders.
t= Q/ D
Note, if the cycle time is greater than the shelf life, items will go bad, and the model must be
modified.

Number of Orders per Year:

To find the number of orders per years take the reciprocal of the cycle time.
n= D / Q = 1 / t
Example: The demand for a product is 1000 units per year. The order size is
250 units under an EOQ policy.
How many orders are placed per year? N = 1000/250 = 4 orders.
How often orders need to be placed (What is the cycle time)?
t = 250 / 1000 = ¼ years. t = ¼ years * 365 days = 91 days
{Note: The four orders are equally spaced}.
Cost Equation for the EOQ Model
TotalAnnual = TotalAnnual + TotalAnnual + TotalAnnual
Inventory Costs Holding Costs Ordering Costs Procurement Costs

TC(Q) = (Q/2)*Ch + (D/Q)*Co + D*Cu

The Optimal Order Size

2DC o
Q* =
Ch
Problem
A two-wheeler component manufacturing unit uses large quantities of a component made
of steel. Although these are production items, the demand is continuous and inventory
planning could be done independent of the production plan. The annual demand for the
component is 2,500 boxes. The company procures the item from a supplier at the rate of
750 per box. The company estimates the cost of carrying inventory to be 18 percent per
unit per annum and the cost of ordering as 1,080 per order. The company works for 250
days in a year. How should the company design an inventory control system for this item?
What is the overall cost of the plan?
 Solution
Annual Demand (D) =2500 boxes.
Number of working days = 250
2500
Average daily demand = = 10 𝑏𝑜𝑥𝑒𝑠/𝑑𝑎𝑦
250
Unit cost of the item = Rs. 750 per box
The company estimates the cost of carrying inventory to be 18 per cent per unit per
annum
 Inventory carrying cost = 𝑖𝑐 = 0.18 × 750 = 𝑅𝑠. 135 𝑝𝑒𝑟 𝑢𝑛𝑖𝑡 𝑝𝑒𝑟 𝑦𝑒𝑎𝑟
 Cost of ordering 𝑐𝑜 = 𝑅𝑠. 1080 𝑝𝑒𝑟 𝑜𝑟𝑑𝑒𝑟.
2𝐷𝐶𝑜 2×2500×1080
 Economic Order Quantity 𝑄∗ = = = 200 𝑏𝑜𝑥𝑒𝑠
𝐶𝑐 135
Solution
𝐷 2500
 Number of order = = = 12.5 ≅ 13
𝑄∗ 200
𝑄∗ 200
 Time between the orders = = = 0.08 𝑦𝑒𝑎𝑟 = 0.08 × 250 𝑑𝑎𝑦𝑠 = 20𝑑𝑎𝑦𝑠.
𝐷 2500
 Total cost of the plan
𝐷 𝑄∗
𝑇𝐶 𝑄 ∗ = ∗ 𝐶0 + 𝐶𝑐
𝑄 2

2500 200
= × 1080 + × 135 = 27,000
200 2
Hence, the manufacturer will place an order for 200 boxes of the component once in
every 20 days and will incur a total cost of 27,000 for the plan. Any quantity above or
below will increase the cost of the plan.
Problem -EOQ
Demand for the computer at XYZ retail store is 1,000 units per month. XYZ incurs a
fixed order placement, transportation, and receiving cost of $4,000 each time an order is
placed. Each computer costs XYZ $500 and the retailer has a holding cost of 20 percent.
Evaluate the number of computers that the store manager should order in each
replenishment lot. How much time each computer spends on average before it is sold?
Solution
• Annual Demand (D) = 1000 × 12 = 12000 𝑢𝑛𝑖𝑡𝑠
• Cost of ordering 𝑐𝑜 = $4,000 𝑝𝑒𝑟 𝑜𝑟𝑑𝑒𝑟
• Unit cost of computer = C =$500
• Holding cost per year as a fraction of unit cost , i =0.2
2𝐷𝐶𝑜 2×12000×4000
• Economic Order Quantity 𝑄∗ = = = 980
𝐶𝑐 0.2×500
𝐷 12000
• Number of orders per year = = = 12.24 ≅ 13
𝑄∗ 980
𝐷 𝑄∗
• Total inventory cost at EOQ = 𝑇𝐶 𝑄∗ = 𝐶 + 𝐶 + 𝐶𝑢 ∗ D = $97,980
𝑄∗ 0 2 𝑐
𝑄 ∗ൗ 980
• Average flow time = 2𝐷 = = 0.041 𝑦𝑒𝑎𝑟 = 0.49 𝑚𝑜𝑛𝑡ℎ
2×12000
Q SYSTEM/ FIXED QTY - VARIABLE INTERVAL
This also known as perpetual inventory system, reorder inventory
system or Q system. In this system, the count of the number of units
in inventory is continuously maintained. With lead time less than
the reorder cycle, an order for a fixed quantity Q (mostly EOQ) is
placed when the inventory level drops to a predetermined reorder
level R.

P SYSTEM / FIXED PERIOD SYSTEM

The system involves the reviewing of stock levels at a fixed interval
of time known as review period and placing replenishment orders at
the end of each period. The replenishment quantity is variable and
corresponds to the amount of stock required to bring the stock
ordered and the stock on hand up to a target inventory level.
 Production Quantity Model
▶An inventory system in which an order is received gradually,
as inventory is simultaneously being depleted
▶AKA non-instantaneous receipt model
▶assumption that Q is received all at once is relaxed
▶p - daily rate at which an order is received over time,
▶ a.k.a. production rate
▶d - daily rate at which inventory is demanded

12-26
 Assumptions continued

• The demand of the item is not high enough to warrant

continuous production. Therefore items are produced in
lots or batches.

12-27
ProductionQuantityModel(cont.)
Inventory
level

Maximum
Q(1-d/p) inventory
level

Average
Q inventory
(1-d/p)
2 level

0
Begin End Time
order order
Order
receipt receipt
receipt period

12-28
ProductionQuantityModel(cont.)

p = production rate d = demand rate

Maximum inventory level = Q - Q d

p

=Q1- d 2CsD
p
Qopt = d
Q Cc 1 - p
Average inventory level = 1- d
2 p

CsD CcQ d
TC = Q + 2 1 - p

12-29
Production Quantity Model: Example
Cc = $0.75 per yard Cs = $150 D = 10,000 yards
d = 10,000/311 = 32.2 yards per day p = 150 yards per day

2CsD 2(150)(10,000)
Qopt = = = 2,256.8 yards
Cc 1 - d 0.75 1 -
32.2
p 150

CsD CcQ d
TC = +
Q 2 1- p = $1,329

Q 2,256.8
Production run = =
p 150 = 15.05 days run
ProductionQuantity Model: Example (cont.)

10,000
Number of production runs = D = = 4.43 runs/year
Q 2,256.8

d
Maximum inventory level = Q 1 - = 2,256.8 1 - 32.2
p 150
= 1,772 yards
Lead Time and the Reorder Point:

In reality lead time always exists, and must be accounted for when deciding
when to place an order.
The reorder point, R, is the inventory position when an order is placed.
R=L*D
L and D must be expressed in the same time unit.
Safety Stock:
Safety stocks act as buffers to handle:
– Higher than average lead time demand.
– Longer than expected lead time.
With the inclusion of safety stock (SS), R is calculated by
R = L * D + SS
The size of the safety stock is based on having a desired service level.
Service Level:
 EOQ Model
Order
O rde quantity,
rder
r Q
quanuantititty, Q Demand
De
Demamandnd
rate
ra
r ate
te

e nto
Inventory
ven Level
ry
t or
velel
Lev
I nv
Le
In

Reorder point, R
ReorderReorder
po
point,t,R
R

0 Lead Lead Time

LeadLead LeadLead
time time
0 Order Order Order Order
placed receipt placed receipt
Lead Time and the Reorder Point: Graphical
demonstration: Short Lead Time

Reorder
Point

Placetheordernow L

R=Inventoryat handatthebeginningofLeadTime
Aggregate Planning
 Lecture Outline

• Hierarchical Nature of Aggregate Planning

• Aggregate Planning Process
• Strategies for Adjusting Capacity
• Strategies for Managing Demand
• Quantitative Techniques for APP
Hierarchical Nature of Planning
Production Capacity Resource
Items Planning Planning Level
Product lines Aggregate Resource
production requirements Plants
or families
plan plan

Individual Master Rough-cut Critical

products production capacity work
schedule plan centers

Material Capacity All work

Components requirements requirements centers
plan plan

Manufacturing Shop Input/ Individual

operations floor output machines
schedule control
 Aggregate Planning
• Collaborative procedure to match demand and capacity
• Translates business demand into marketing plans into production
plans for all product families
• Medium range plans (3 to 18 months)
• Within this time frame, it is usually
• not feasible to increase capacity by building new facilities or
purchase new equipments;
• however it is feasible to hire or lay off workers, increase or
reduce the workweek, add an extra shift, subcontract the work,
use overtime or flexi-time, build up or deplete the inventory
levels.
 Aggregate Planning
• Also known a Sales and Operations Planning (SO&P)
• Determine the resource capacity needed to meet demand over an
intermediate time horizon
• Aggregate refers to product lines or families
• Resource capacity is exp’d in aggregate terms
• Labour hours, machine hours, space, time etc.
• Aggregate planning matches capacity and demand
• Objectives
• Establish a company wide game plan for allocating resources
• Develop an economic strategy for meeting demand
• The objective is to minimize the cost of resources required to meet the
demand over that period.
Strategies for Adjusting Capacity
• Peak Demand
• Inventory based
• Varying workforce levels
• Over-time and Under-time
• Subcontracting
• Part-time Workers
Strategies for Adjusting Demand

• Stimulating/Influencing demand
• Backordering during high demand period
• Counter seasonal products/services
Strategies for Adjusting Capacity
• Level Production
• Producing at a constant rate, usually at average
demand rate, and using inventory to absorb
fluctuations in demand.
• During periods of low demand, over-production is
stored as inventory
• Periods of high demand are taken care off through
the stored inventory
• Cost of this strategy is the cost of holding inventory,
including the cost of obsolete and perishable items
 Level Production
Demand

Production

Units

Time
 Strategies for Adjusting Capacity
• Chase Demand
• Matches the production plan to the demand pattern and absorbs variation
by hiring and firing workers
• During the periods of high demand workers are hired to increase the
production
• The cost of this strategy is hiring and firing the workers
• Cost effective during the periods of high unemployment or for industries
with low skill requirements
• Backfires in case of industries which needs higher level of skills and where
the labour is scarce and competition for labour is intense
Chase Demand

Demand

Production
Units

Time
 Mixed Strategy
• Combination of Level Production and Chase Demand
strategies
• Examples of management policies
• no more than x% of the workforce can be laid off in one
quarter
• inventory levels cannot exceed x rupees
• production levels should not go down x levels of pre-
specified trade-offs between important factors
• Many industries may simply shut down manufacturing
during the low demand season and schedule employee
vacations during that time
Quantitative Techniques
• Pure Strategies
• Mixed Strategies
• Transportation Method
• Other Quantitative Techniques
• The most effective strategy depends on

• Demand distribution
• Competitive position
• Cost structure of the firm or product line
 Pure Strategies
• ABC candy company makes a variety of candies in three factories world
wide. Its line of chocolate candies exhibit a highly seasonal pattern, with
peaks during the winter months and troughs during the summers. Given the
following costs and quarterly sales forecasts, determine whether [a] level
production, or [b] chase demand would more economical to meet the
demand for chocolate candies:
Pure Strategies

Example: QUARTER SALES FORECAST (KG)

I 80,000
II 50,000
III 120,000
IV 150,000
Hiring cost = Rs. 100 per worker
Firing cost = Rs. 500 per worker
Regular production cost per kg = Rs. 2.00
Inventory carrying cost = Rs. 0.50 per quarter
Production per employee = 1,000 kg. per quarter
Beginning work force = 100 workers
Level Production Strategy
Level production
(50,000 + 120,000 + 150,000 + 80,000)
= 100,000 KGS
4
SALES PRODUCTION
QUARTER FORECAST PLAN INVENTORY
I 80,000 100,000 20,000
II 50,000 100,000 70,000
III 120,000 100,000 50,000
IV 150,000 100,000 0
400,000 140,000
Cost of Level Production Strategy
(400000 X 2.00) + (140000 X 0.50) = Rs. 8,70,000
Chase Demand Strategy

SALES PRODUCTION WORKERS WORKERS WORKERS

QUARTER FORECAST PLAN NEEDED HIRED FIRED

I 80,000 80,000 80 0 20
II 50,000 50,000 50 0 30
III 120,000 120,000 120 70 0
IV 150,000 150,000 150 30 0
100 50

Cost of Chase Demand Strategy

(400,000 X 2.00) + (100 x 100) + (50 x 500) = Rs. 8,35,000
 Transportation Method
EXPECTED REGULAR OVERTIME SUBCONTRACT
QUARTER DEMAND CAPACITY CAPACITY CAPACITY

1 900 1000 100 500

2 1500 1200 150 500
3 1600 1300 200 500
4 3000 1300 200 500

Regular production cost per unit Rs. 20

Overtime production cost per unit Rs. 25
Subcontracting cost per unit Rs. 28
Inventory holding cost per unit per period Rs. 3
Beginning inventory 300 units
No backordering is permitted. Design a production plan that will
satisfy the demand at lowest possible cost.
 Period of Use Units Unused

Period of Production 1 2 3 4 Produced Capacity Capacity

Beg. Inventory 300 0 0 3 0 6 9 300

1 Regular 20 23 26 29 1,000

Overtime 25 28 31 34 100

Subk 28 31 34 37 500

2 Regular 20 23 26 1,200

Overtime 25 28 31 150

Subk 28 31 34 500

3 Regular 20 23 1,300

Overtime 25 28 200

Subk 28 31 500

4 Regular 20 1,300

Overtime 25 200

Subk 28 500

Units Supplied 7,000

Demand 900 1,500 1,600 3,000 7,000

Unmet Demand Total Cost =
 Period of Use Units Unused

Period of Production 1 2 3 4 Produced Capacity Capacity

Beg. Inventory 300 0 0 3 0 6 0 9 300 300

1 Regular 600 20 23 26 29 1,000

Overtime 25 28 31 34 100

Subk 28 31 34 37 500

2 Regular 1200 20 23 26 1,200

Overtime 25 28 31 150

Subk 28 31 34 500

3 Regular 1300 20 23 1,300

Overtime 25 28 200

Subk 28 31 500

4 Regular 1300 20 1,300

Overtime 25 200

Subk 28 500

Units Supplied 7,000

Demand 900 1,500 1,600 3,000 7,000

Unmet Demand 0 0 0 0 Total Cost =
 Period of Use Units Unused

Period of Production 1 2 3 4 Produced Capacity Capacity

Beg. Inventory 300 0 0 3 0 6 0 9 300 300 0

1 Regular 600 20 300 23 100 26 0 29 1,000 1,000 0

Overtime 0 25 0 28 0 31 100 34 100 100 0

Subk 0 28 0 31 0 34 0 37 0 500 500

2 Regular 0 1200 20 0 23 0 26 1200 1,200 0

Overtime 0 0 25 0 28 150 31 150 150 0

Subk 0 0 28 0 31 250 34 250 500 250

3 Regular 0 0 1300 20 0 23 1,300 1,300 0

Overtime 0 0 200 25 0 28 200 200 0

Subk 0 0 0 28 500 31 500 500 0

4 Regular 0 0 0 1300 20 1,300 1,300 0

Overtime 0 0 0 200 25 200 200 0

Subk 0 0 0 500 28 500 500 0

Units Supplied 900 1,500 1,600 3,000 7,000 7,750 750

Demand 900 1,500 1,600 3,000 7,000

Unmet Demand 0 0 0 0 Total Cost = $153,550
 Production Plan
REGULAR SUB- ENDING
PERIOD DEMAND PRODUCTION OVERTIME CONTRACT INVENTORY

1 900 1000 100 0 (OS:3+2)500

2 1500 1200 150 250 (5+1)600

3 1600 1300 200 500 (6+4) 1000

4 3000 1300 200 500 0
-------------------------------------------------------------------------
Total 7000 4800 650 1250 2100
Cost 20 25 28 3
-------------------------------------------------------------------------
Total Cost = 1,53,550 Rs.
 AP for Services
1. Most services can’t be inventoried
2. Demand for services is difficult to predict
3. Capacity is also difficult to predict
4. Service capacity must be provided at the
appropriate place and time
5. Labor is usually the most constraining resource for
services
Material Requirement Planning
Material Requirements
Planning (MRP)
• MASTER PRODUCTION SCHEDULE
• DEPENDENT DEMAND
• PRODUCT STRUCTURE
• BILLS OF MATERIALS
• MASTER INVENTORY FILES

Lot Sizing MRP Rules

• LOT FOR LOT
• EOQ
Material Requirements Planning (MRP)

Computerized inventory control (r/m, components, f/g) &

production planning system (purchase, produce, deliver)
Objectives
• Ensure that the material is available when needed (funds)
To maintain lowest possible level of inventory (when)

When to use MRP?

• Dependent demand items: derived
• Discrete demand items: components, lumps, EOQ, issues
Complex products: multi-layered, tiered, sub-assemblies
Demand Characteristics
Material Requirements Planning
Master Production Schedule

0 60 0 60 0
Product Structure File
Product Structure Tree
 Bills of Material (BOM)

1. Product Structure File is a file that contains a computerized

bills of material (BOM) for every item produced.
2. Bills of Material for a product lists the items that go into the
product, includes a brief description of each item, and
specifies when and in what quantity each item is needed in
the assembly process.
 MRP Processes
The MRP system is responsible for scheduling the
production of all items beneath the end item level.

•Exploding the bill of material: the process of

determining requirements of lower level items
• Netting out inventory: the process of subtracting on-
hand quantities and scheduled receipts from gross
requirements to produce net requirements.
• Lot sizing: determining the quantities in which the
items are usually made or purchased
• Time-phasing requirements: subtracting an item’s lead
time from its due date to determine when to order an
item
MRP Matrix
MRP: Example (cont.)
MRP: Example (cont.)
 MRP: Example (cont.)

115 units available

(115 - 95) = 20 on hand at the end of Period 2
MRP: Example (cont.)
MRP: Example (cont.)
MRP: Example (cont.)
MRP: Example (cont.)
MRP: Example (cont.)
MRP: Example (cont.)
MRP: Example (cont.)
MRP: Example (cont.)
FORECASTING
 Forecasting

 Predicting the Future

 Vital for business organization
 Underlying basis of all business decisions
 Most techniques assume an underlying stability in the
system
 Qualitative Forecasting Approach:
 Quantitative Forecasting Approach:
 Qualitative Methods
• Grass root method – going down to the lowest level of
hierarchy
• Market research – data collection and hypothesis testing
• Jury of ‘executive opinion’ – source of internal qualitative
forecast
• Historical analogy – history or past data of the item
• Panel consensus – free open exchange in between select few
• Delphi Method - Iterative group process.
• 3 types of participants
• Decision makers: Evaluate responses and make decisions
• Staff: Administering survey
• Respondents: People who can make valuable judgments
 Quantitative Forecasting
Time Series Models:
• Set of evenly spaced numerical data - Obtained by observing
response variable at regular time periods.
• Forecast based only on past values - Assumes that factors
influencing past and present will continue influence in future
1. Naive approach
2. Moving averages
3. Exponential smoothing
4. Trend projection

Associative Models / Causal Models:

1. Linear regression
 Naive Approach
ORDERS
MONTH PER MONTH FORECAST

Jan 120 -
Feb 90 120
Mar 100 90
Apr 75 100
May 110 75
June 50 110
July 75 50
Aug 130 75
Sept 110 130
Oct 90 110
Nov - 90
Simple Moving Average
5 Month Simple Moving Average

ORDERS MOVING
5
MONTH PER MONTH AVERAGE
Jan 120 –

i=1
Di

Feb 90 – MA5 =
5
Mar 100 –
Apr 75 –
May 110 – 120 + 90 + 100+75+110
=
June 50 99.0 5
July 75 85.0
Aug 130 82.0 = 99 orders for June.
Sept 110 88.0
Oct 90 95.0
Nov - 91.0
8
3 Month Simple Moving Average
ORDERS MOVING
MONTH PER MONTH AVERAGE
3
Jan 120 –
 Di
Feb 90 – i=1
MA3 =
Mar 100 – 3
Apr 75 103.3
May 110 88.3 120 + 90 + 100
= 3
June 50 95.0
July 75 78.3
Aug 130 78.3 = 103.3 orders for Apr.
Sept 110 85.0
Oct 90 105.0
Nov - 110.0
 Weighted Moving Average
 Adjusts moving average method to more closely reflect data fluctuations
 Weights are assigned to most recent data, barring in case of seasonal cycles
 Precise weights are decided thorough trial and error (based on experience
and intuition), as does the number of periods to be considered
 If recent periods are weighted too heavily, the forecast might over-react to a
random fluctuation in demand
 If they are weighted too lightly, the forecast might under-react
to actual changes in demand pattern
Weighted Moving Average

WMAn = WD i i

i=1
where
Wi = the weight for period i,
between 0 and 100 percent

 Wi = 1.00
Weighted Moving Average

MONTH WEIGHT DATA

August 17% 130
September 33% 110
October 50% 90
3
November Forecast WMA3 =  Wi Di
i=1

= (0.50) (90) + (0.33) (110) + (0.17) (130)

= 103.4 orders
11
 Exponential Smoothing
 Averaging method - weights most recent data more strongly
 As the past becomes more distant, the imp. of data diminishes
 So very useful and preferable method, if recent changes are significant and
unpredictable
 Widely used, most popular because its an accurate method
 Requires minimal data:
forecast for the current period,
actual demand for the current period and
a weighing factor OR smoothing constant.
 Has good track record of success
 Found to be used and preferred method by most companies
Exponential Smoothing
Ft+1 =   Dt + (1 - ) * Ft
where:
Ft + 1 =forecast for next period
Dt = actual demand for present period
Ft = previously determined forecast for present period
 = weighting factor, smoothing constant –
determines the level of smoothing
*Assume first forecast as Actual Demand…
Effect of Smoothing Constant
0.0    1.0
reflects the weight given to the most recent demand data
If  = 0.20, then Ft + 1 = 0.20 * Dt + 0.8 * Ft
If  = 0, then Ft + 1 = Ft
Forecast does not even consider recent actual data
If  = 1, then Ft + 1 = 1 * Dt + 0 * Ft = Dt
Forecast based only on most recent data, so this becomes
as good as naïve forecast
Exponential Smoothing (α = 0.30)

PERIOD MONTH DEMAND

1 Jan 37
F2 =  D1 + (1- ) F1
2 Feb 40
3 Mar 41 = (0.30) 37 + (1- 0.3) 37
4 Apr 37 = 37
5 May 45
6 Jun 50 F3 =  D2 + (1- ) F2
7 Jul 43
= (0.30) 40 + (1- 0.3) 37
8 Aug 47
9 Sep 56 = 37.90
10 Oct 52
11 Nov 55 F13 =  D12 + (1- ) F12
12 Dec 54 = (0.30) 54 + (1- 0.3) 50.84
= 51.79
 Exponential Smoothing
FORECAST, Ft + 1
PERIOD MONTH DEMAND ( = 0.3) ( = 0.5)

1 Jan 37 – –
2 Feb 40 37.00 37.00
3 Mar 41 37.90 38.50
4 Apr 37 38.83 39.75
5 May 45 38.28 38.37
6 Jun 50 40.29 41.68
7 Jul 43 43.20 45.84
8 Aug 47 43.14 44.42
9 Sep 56 44.30 45.71
10 Oct 52 47.81 50.85
11 Nov 55 49.06 51.42
12 Dec 54 50.84 53.21
13 Jan – 51.79 53.61 16
 Regression Methods
 Linear Regression
 Regression can be defined as functional relationship between two or
more correlated variables
 Regression is used for forecasting by establishing a mathematical
relationship between two or more variables (demand and some
other independent variable) in the form of a linear equation
 It is used to predict one variable given the other
 Linear regression refers to the special class of regression where the
relationship between the variable forms a straight line
 Good for long range forecasting and aggregate planning

18
 Linear Trend Line
Linear Regression is a causal
method of forecasting in which a
mathematical relationship is
developed between demand and
time.  xy - nxy
b =
 x2 - nx2
Linear trend line relates a
dependent variable (demand) to an a = y - bx
independent variable (time) in the
where
form of a linear equation: n = number of periods
y = a + bx
x
x =
a = intercept n = mean of the x values
b = slope of the line  y = mean of the y values
x = time period y =
n
y = demand forecast for period x
19
 Least Squares Example
x (PERIOD) y (DEMAND) xy x2
1 37 37 1
2 40 80 4
3 41 123 9
4 37 148 16
5 45 225 25
6 50 300 36
7 43 301 49
8 47 376 64
9 56 504 81
10 52 520 100
11 55 605 121
12 54 648 144
78 557 3867 650
20
 Least Squares Example

x = 78 = 6.5
12
y = 557 = 46.42
12
b = xy - nxy = 3867 - (12)(6.5)(46.42) =1.72
x2 - nx2 650 - 12(6.5)2
a = y - bx
= 46.42 - (1.72)(6.5) = 35.2

21
 Linear Trend Line y = 35.2 + 1.72x
Forecast for Period 13 y = 35.2 + 1.72(13) = 57.56 units

70 –

60 – Actual

50 –
Demand

40 –
Linear trend line

30 –

20 –

| | | | | | | | | | | | |
10 – 6 7
1 2 3 4 5 8 9 10 11 12 13
Period
0– 22
 Linear Regression Example
x y
adv spend sales xy x2

4 36.3 145.2 16
6 40.1 240.6 36
6 41.2 247.2 36
8 53.0 424.0 64
6 44.0 264.0 36
7 45.6 319.2 49
5 39.0 195.0 25
7 47.5 332.5 49
49 346.7 2167.7 311
Linear Regression Example (cont.)
49
x= = 6.125
8
346.9
y= = 43.36
8

xy - nxy
b=
x2 - nx2
(2,167.7) - (8)(6.125)(43.36)
= (311) - (8)(6.125)2
= 4.06

a = y - bx
= 43.36 - (4.06)(6.125)
= 18.46
Linear Regression Example (cont.)
Correlation & Coefficient of Determination
• Correlation, r
• Correlation is a measure of the strength of the relationship between
independent and dependent variables
• degree of association between two variables (-1.00 to +1.00)
• nil/poor/average/strong, & positive/negative

• Coefficient of Determination, r2
• Percentage of variation in dependent variable resulting from changes in the
independent variable (0% to 100%)
• A measure of the amount of variation in the dependent variable about
its mean that is explained by the regression equation
 Computing Correlation

n xy -  x y
r=
[n x2 - ( x)2] [n y2 - ( y)2]
(8)(2,167.7) - (49)(346.9)
r=
[(8)(311) - (49)2] [(8)(15,224.7) - (346.9)2]

r = 0.947

Coefficient of Determination
r2 = (0.947)2 = 0.897
27
 Seasonal Adjustments
 Repetitive increase / decrease in demand
 Seasonal patterns can also occur on a periodic basis
 Use seasonal factor to adjust forecast
 A seasonal factor is a numeric value that is multiplied by the normal
forecast to get a seasonally adjusted forecast
 A seasonal factor range from 0 to 1, it is in effect, the portion of
annual demand assigned to each season
 Thus SF when multiplied to annual forecasted demand yield
seasonally adjusted forecasts for each season
Di
Seasonal Factor = S= i
D
8
 Seasonal Adjustment (cont.)
DEMAND (1000’S PER QUARTER)
YEAR I II III IV Total
2002 12.6 8.6 6.3 17.5 45.0
2003 14.1 10.3 7.5 18.2 50.1
2004 15.3 10.6 8.1 19.6 53.6
Total 42.0 29.5 21.9 55.3 148.7

D1 42.0 D3 21.9
SI = SIII = D = 148.7 = 0.15
D = 148.7= 0.28
D2 29.5 D4 55.3
SII = D = = 0.20 SIV = D = = 0.37
148.7 148.7
Seasonal Adjustment (cont.)
 X Y X*X X*Y
 1 45.0 1 45.00
 2 50.1 4 100.20
 3 53.6 9 160.80

 FIND MEAN OF X AND Y

 VALUE OF a AND b
Seasonal Adjustment (cont.)
For 2005

y =40.97 +4.30 x =40.97 +4.30(4) =58.17

SFI =(SI) (F5) =(0.28)(58.17) =16.28

SFII =(SII) (F5) =(0.20)(58.17) =11.63
SFIII =(SIII) (F5) =(0.15)(58.17) =8.73
SFIV =(SIV) (F5) =(0.37)(58.17) =21.53
 Forecast Accuracy
 A forecast isnever ever accurate
 Large degree of error mean
 Either the forecasting technique used is applied wrongly or is not
applicable in the case
 Wrong relationship among variables
 Or the ‘parameters’ used need to be adjusted for ‘trend’
 Forecast Error
 Difference between forecast and actual demand - Error
 MAD - Mean Absolute Deviation
 MAPD - Mean Absolute Percent Deviation or MAPE
 Cumulative Error - RSFE
 Average Error or Bias
 Mean Absolute Deviation (MAD)

  Dt - Ft 
MAD = n
MAD: The absolute average difference between the AD & FD.
where,
t =period number
Dt =demand in period t
Ft =forecast for period t
n =total number of periods
=absolute value
The smaller the value of MAD relative to the magnitude of the data the
accurate the forecast.
 Other Accuracy Measures
 MAPD: Measures the absolute error (AV-FV) as a % of demand
rather than per period (MAD). Can be used across the board to
measure the relative accuracy of the forecast.
 Cumulative Error (RSFE): Simply computed by summing up
the forecast errors. That’s why Linear Trend Line has zero
cumulative value.
 Average Error (Bias): Computed by averaging the cumulative
error value (RSFE) over the number of time periods.
 +ve value: low, -ve value: high and zero value: no bias
Other Accuracy Measures

Mean Absolute Percent Deviation (MAPD)

 |Dt - Ft|
MAPD =
Dt
Cumulative Error (RSFE)
RSFE =  et =  (Dt – Ft)
Average Error (Bias)
 et
E= n 35
 MAD Example
PERIOD DEMAND, Dt Ft ( =0.3) (Dt - Ft) |Dt - Ft|
1 37 37.00 – –
2 40 37.00 3.00 3.00
3 41 37.90 3.10 3.10
4 37 38.83 -1.83 1.83
5 45 38.28 6.72 6.72
6 50 40.29 9.69 9.69
7 43 43.20 -0.20 0.20
8 47 43.14 3.86 3.86
9 56 44.30 11.70 11 .70
10 52 47.81 4.19 4.19
11 55 49.06 5.94 5.94
12 54 50.84 3.15 3.15
557 49.31 53.39
MAD Example
  Dt - Ft 
MAD = n
53.39
=
11
= 4.85
 Forecast Control
 Forecast can go out of control due to various reasons
 Change in trend
 Unanticipated appearance of a cycle
 Irregular variation such as unseasonable weather
 Promotional campaign, new competition, political
reasons, others…
 Tracking Signal: this indicates whether the forecast average is
keeping pace with any
genuine upward or downward changes in demand
 Monitors the forecast to see if it is biased high or low
 Control limits of 2 to 4 MADs are used most frequently

(Dt - Ft) MAD

Tracking Signal = MAD = RSFE
 Tracking Signal Values
DEMAND FORECAST, ERROR E = TRACKING
PERIOD Dt Ft Dt - Ft (Dt - Ft) MAD SIGNAL

1 37 37.00 – – – –
2 40 37.00 3.00 3.00 3.00 1.00
3 41 37.90 3.10 6.10 3. 05 2.00
4 37 38.83 -1.83 4.27 2.64 1.62
6.10
5 45 38.28 6.72 10.99 3.66 3.00 TS3 = = 2.00
6 50 40.29 9.69 20.68 4.87 4.25 3.05
7 43 43.20 -0.20 20.48 4.09 5.01
8 47 43.14 3.86 24.34 4.06 6.00
9 56 44.30 11.70 36.04 5.01 7.19
10 52 47.81 4.19 40.23 4.92 8.18
11 55 49.06 5.94 46.17 5.02 9.20
12 54 50.84 3.15 49.32 4.85 10.17
Example

41
Example
 Forecasting Process
1. Identify the 2. Collect historical 3. Plot data and
purpose of forecast data identify patterns

6. Check forecast 5. Develop/compute 4. Select a forecast

accuracy with one forecast for period model that seems
or more measures of historical data appropriate for data

7.
Is accuracy No 8b. Select new
of forecast forecast model or
acceptable? adjust parameters
of existing model
Yes

9. Adjust forecast 10. Monitor results

8a. Forecast over
based on additional and measure
planning horizon
qualitative info’ & insight forecast accuracy

43