Professional Documents
Culture Documents
SYNOPSIS
1. Top-Down Approach:
The top-down approach develops a national level forecast and then
spreads the volume across locations on the basis of historical sales
pattern. For example, suppose the aggregate monthly forecast for the
entire country is 10000 units. Assume that the firm uses four
distribution centers to serve the demand with the historical split of 40,
30, 20 and 10 percent, respectively. In this case, forecasts for individual
distribution centers are projected to be 4000, 3000, 2000 and 1000
units respectively.
56 Logistics and S C M (TYBMS Sem – V)
2. Bottom-Up Approach:
On the other hand the bottom-up approach is decentralized since each
distribution center forecast is developed independently. As a result,
each forecast can more accurately track and consider demand
fluctuations within specific markets. However, the bottom up approach
requires more detailed record keeping.
Example: In bottom-up approach, each distribution centre will forecast
its demand individually. Addition of this individual demands will give
overall corporate demand.
ensure whether they are realistic or not and later they are
combined to arrive at a forecast.
4. Naive Approach:- Naive Approach to demand forecasting is
based on the assumption that demand in the next period is similar
to the demand in most recent period without major changes in
demand patterns.
5. Delphi Method:- In Delphi method, an attempt is made to
develop forecasts through “group consensus”. This technique
relies on a panel of experts. Each expert is asked to answer
questionnaires in two or more rounds.
After each round, a facilitator provides a small summary of each
experts forecasts from the previous round as well as the reasons
they provided for their judgment. Each expert is asked to revise
and to reconsider their earlier answers. It is believed that during
this process the range of answers will decrease and group will
arrive at a consensus. The process is stopped after a pre-defined
stop criterion. The criteria may be number of rounds, achievement
of consensus, etc.
B. Quantitative Methods of Forecasting
Quantitative methods are based on analysis of historical data. It is objective
in nature. Quantitative techniques include the following methods:
1. Time Series Method:
Time Series Method is based on basic assumption that the future
will be similar to the past. This implies that existing demand
patterns will continue in the future.
This assumption is often reasonably correct in the short term.
Thus, these techniques are most appropriate for short-range
forecasting.
When the rate of growth or the trend changes significantly, the
demand pattern experiences a turning point. Since time-series
techniques use historical demand patterns, they are typically not
sensitive to turning points.
Time Series method includes:
1. Moving Average: Moving average forecasting uses an average of the
most recent period’s sales. The average may contain any number of
previous time periods. Three, four and five period averages are
commonly used.
The moving averages are easy to calculate but there are some
limitations. They do not respond to sudden market changes. Moreover,
a large amount of historical data must be maintained and updated to
calculate forecasts.
58 Logistics and S C M (TYBMS Sem – V)
Solution:
Years Sales 2 Yearly Moving 3 Yearly Moving
(in 000’s) Averages Averages
(in ‘000 Rs.) (in ‘000 Rs.)
1995 45 - -
1996 52 - -
1997 83 48.5
1998 92 67.5 60
1999 98 87.5 75.67
2000 115 95 91
2001 125 106.5 101.67
120 112.67
60 Logistics and S C M (TYBMS Sem – V)
Solution:
Solution:
Years Sales 2 Period Moving Averages 5 Period Moving Averages
(in 000’s) (in ‘ 000 Rs.) (in ‘000 Rs.)
2000 180 - -
2001 195 - -
2002 200 187.5 -
2003 211 197.5 -
2004 222 205.5 -
2005 240 216.5 201.6
2006 269 231 213.6
2007 284 254.5 228.4
2008 300 276.5 245.2
2009 305 292 263
2010 315 302.5 279.5
2011 322 310 294.6
2012 330 318.5 305.2
2013 335 326 314.4
2014 340 332.5 321.4
2015 350 337.5 328.4
2016 345 335.4
Q.4. You are given the following information about demand of an
item.
Months Demand
1 100
2 150
3 225
4 240
5 170
6 280
7 300
8 300
9 320
10 330
11 350
Calculate:
i. 3 – monthly moving averages
ii. 4 - monthly moving averages
iii. Also calculate 4 - monthly weighted moving averages with weights as
4:3:2:1. the largest weight being for the most recent value.
iv. What will be the forecast for 12th month.
62 Logistics and S C M (TYBMS Sem – V)
Solution:
Q.5. The demand for an item in the past few years is as follows:
Years Demand
2006 1100
2007 1200
2008 1400
2009 1500
2010 1800
2011 2000
2012 2100
2013 2200
2014 2300
2015 2400
Calculate:
i. Calculate 3 - yearly moving average.
ii. Calculate 4 - yearly moving average.
iii. Calculate 3 yearly weighted moving average with weights as 3:2:1, the
largest weight being for the most recent value.
iiv. Also calculate forecast for the year 2016.
Solution:
Solution:
1 160 - -
2 180 - -
3 190 - -
4 210 (190 x 3) + (180 x 2) + (160 x 1) 1090 182
=
3+2+1 6
5 230 (210 x 3) + (190 x 2) + (180 x 1) 1190 198
=
3+2+1 6
6 240 (230 x 3) + (210 x 2) + (190 x 1) 1300 217
=
3+2+1 6
7 250 (240 x 3) + (230 x 2) + (210 x 1) 1390 232
=
3+2+1 6
8 (250 x 3) + (240 x 2) + (230 x 1) 1460 243
=
3+2+1 6
Solution:
REVIEW QUESTIONS