FORECASTING SALES OF SIM CARD “SIMPATI”
IN PT. SIMPATINDO MULTIMEDIA BRANCH OFFICE PENGGILINGAN CAKUNG, EAST JAKARTA”
4EA28
Aliyaturrohmah 10212651
Desy Purnamasari 11212910
Endang Suliswati 12212347
Laila Majda 14212153
Lelly Octavia 14212168
Munawarah Zulhijah 15212158
Nurilita Wiguna 15212500
Puput Novel 1A213858
Wahyu Kusuma Dewi 17212638
GUNADARMA
UNIVERSITY
2016
CHAPTER
I
PRELIMINARY
1.1
Background
Currently, the development of technology in
Indonesia is very rapid. One of them is her first cell phone use cell phones
are expensive items that can only be used by certain circles, in Indonesia the
use of mobile phones very rapidly in the last five years. Before 2008 mobile
phone users in Indonesia is less than 80% of the population, and continues
rapidly developing States from 2009 until today.
With the increasing use of mobile phones increases
then also use SIM cards or also commonly known as the sim card. Along with the
increasing use of SIM cards in Indonesia resulted in the level of competition
in the mobile carrier companies are becoming increasingly stringent. With so
many mobile phone operators for market share in Indonesia.
Therefore the company should be able to take the
right decision for the company to keep it running properly. One of them know
how much product sales. So with that picture, the company can menentukkan
various decisions in order to anticipate the situation is not expected in the
foreseeable future.
To be able to take a decision, we need a careful
planning in order to set objectives will be achieved. This is consistent with
the role of planning is a basic process management in taking a decision and
action. In a company selling a product is a very important aspect for the
survival of the company. Therefore the preparation of sales planning needs to
be done carefully and accurately to determine the development of the company.
So in this case need to do a sales forecast that can help the company in terms
of product sellers in the future.
Proper use of quantitative methods is important,
because not all methods can be used for every issue. Therefore, we must choose
a suitable method. Based on these descriptions, have prompted the authors to
take the title "FORECASTING SALES OF SIM CARD “SIMPATI” IN PT. SIMPATINDO
MULTIMEDIA BRANCH OFFICE PENGGILINGAN CAKUNG, EAST JAKARTA”
1.2
Formulation
and Limitations
1.2.1
Problem
Formulation
The formulation of the problem in this research is:
What is forecasting sales of SIM Card Simpati for the month of August 2015 and
with Which method is better used by enterprises in the forecast sales of SIM
cards?
1.2.2
Limitations
Scientific writing is restricted to sales
forecasting calculation problem SIM cards Sympathy from the date of January 1,
2013 - July 15, 2015 and the data was taken on July 16, 2015.
1.3
Research Purposes
The purpose of this paper is to determine the sales forecasting Prime Card
Sympathy for the month of August 2015 using the Moving Average Weight Moving
Average, Exponential Smoothing and with Which method is better used by enterprises
in the forecast sales of SIM cards in order to minimize human error.
1.4
Benefits of Research
Benefits of
this paper is:
1.
Academic
Benefits
Adding to
the experience and conduct scientific research and be able to apply the
theories that have been the authors obtained in college and compared in the
field.
2.
Practical
Benefits
As a means
for the development of knowledge and insight into the thinking as a scientific
information and the development of knowledge about the product sales
forecasting.
1.5
Research Methods
1.5.1
Reseach Object
The
object of this study is PT.SIMPATINDO MULTIMEDIA grinding branch is located at
Jalan Raya Milling rt 04 / rw 011 Cakung, East Jakarta.
1.5.2
Data /Variable
The
data used in this paper is secondary data obtained from PT.Simpatindo Multimedia
SIM Card sales data in the form of sympathy from the date of January 1, 2013 -
July 15, 2015 will be used as research material.
1.5.3
Data Collection Methods
In
an effort to obtain the data used to answer the problems that have been
identified, the authors tried to collect data through various means:
1.
Research
Library, using references from a variety of books related to the material that
will be discussed in scientific writing.
2.
Field
Research, the research used direct observation by reviewing the company in
order to obtain data that is more objective. Data scientific writing is
obtained by conducting interviews with the authorities in providing an
explanation.
1.5.4
Analysis Tools Used
Analysis tools used in scientific writing is
forecasting using Moving Average Weight Moving Average, Exponential Smoothing
CHAPTER
II
THEORETICAL BASIS
2.1
Theoretical Framework
2.1.1
Definition of Forecasting
Definition of demand
forecasting for its
products is vital in the planning and monitoring of a business entity.
According to Pangestu Subagyo
(2002: 25): "Forecasting is a prediction or forecast that has not
happened. In the social sciences everything is uncertain, it is difficult to
accurately predict, therefore the use of forecasting which aims to forecast or
prediction that can be made to minimize the effect of this uncertainty on the
company ".
According to Jay Heizer and Barry
Render (2006: 136): "Forecasting is the art or science to predict the
future."
2.1.2
Definition of
Sales
According to Mulyadi (2008:
202),: "Sales is an activity performed by the seller sells goods or
services in the hope of getting profit from the existence of such transactions
and sales can be interpreted as a diversion or transfer of ownership of goods
or services from the seller to the buyer. "
According to Kotler (2006: 457):
"Sales is a process whereby the needs of buyers and sales requirements are
met, through an exchange of information and interests".
2.1.3
Definition of
Sales Forecasting
According to Sofyan Assauri (1991;
108): "Sales forecasting is an estimate of quantitative traits, including
the price of the development of the market of a product produced by the company
at a certain period in the future".
From the above definition can be concluded that the
sales forecasting is to estimate the needs of buyers who will come to extend to
a maximum benefit.
2.1.4
Forecasting Horizon Time
Forecasting is usually classified by the future time
horizon that is divided into several categories according to Jay Heizer and
Barry Render (2006: 137) :
1.
Short-term Forecasting
This
forecasting includes a period of up to one year, but generally less than 3
months. Forecasting is used to plan purchasing, work scheduling, workforce, and
production levels.
2.
Medium-term Forecasting
This
forecasting generally covers the monthly count to 3 years. This forecasting is
used to plan sales, planning and production budgets, the cash budget, and
analyze a variety of operations plan.
3.
Long-term Forecasting
This
forecasting is generally for the planning period of 3 years or more. This
forecasting is used to plan new products, capital expenditures, location or
facility development, and research and development.
Medium-term and long-term forecasting can be distinguished from
short-term forecasting with
see three things :
1.
Medium-term
and long-term forecasting deals with the issues more thoroughly and to support
management decisions related to product design, manufacturing, and process.
Setting a decision will be facilities, such as the decision of a general
manager to open a new manufacturing plant in Brazil, it can take 5-8 years from
the beginning to really - really finished completely.
2.
Short-term
Forecasting usually apply a different methodology than the long-term
forecasting. Mathematical techniques, such as average - moving average,
exponential smoothing, and
trend extrapolation
is generally known for short-term forecasting.
3.
Short-term
Forecasting tend to be more precise than the long-term forecasting. Affecting factors changes in demand changed daily.
Thus, in line with the longer the time horizon, the forecasting accuracy of a
person tends to wane. Sales forecasting must be updated regularly to maintain
the value and integrity. Forecasting
should always be reviewed and revised at each end of the sales period.
Forecasting
consists of seven basic steps:
1.
Establish
the purpose of forecasting
2.
Choose
what elements would be predictable.
3.
Determine
the time horizon of the forecast.
4.
Choosing
the type of forecasting models.
5.
Collect
the data needed to conduct forecasting.
6.
Make
forecasting.
7.
Implement
forecasting results
2.1.5
Type
of Forecasting
Organizations typically use three main types of
forecasting in the planning of future operations by Hery Prasetya and Fitri
Lukiastuti (2009: 44):
1.
Economic Forecasting
Is forecasting that explain the
business cycle to predict the rate of inflation, the availability of money, the
funds needed to build housing and other planning indicators. This forecasting
is planning a useful indicator helps organizations to prepare a medium-term
forecasting to long term.
2.
Technology Forecasting
Is
forecasting that takes into account the rate of technological progress can
launch exciting new products, which require new factories and equipment. This
forecasting This usually requires a long period of time by observing the rate
of technological progress.
3.
Demand Forecasting
Is the projection of
demand for the products or services of a company. This forecasting is also
called sales forecasting, which controls production, capacity and scheduling
system and becomes an input for financial planning, marketing and human
resources. This forecasting is predicting sales of a company at any period in
time horizon.
2.1.6
Forecasting Process
Forecasting process according T.Hani Handoko (2000:
260) usually consists of steps - steps as follows:
1.
Determination
of Interest
The
first step consists of determining the estimate of the desired kind. Instead,
depending on the destination information needs of managers. Analysis discuss
with decision makers to know what their needs and determine:
a.
Variables
- what variables to be estimated.
b.
Who
will use the results of forecasting.
c.
For
the purpose - what purpose forecasting results will be used.
d.
Estimated
long-term or short-term desired.
e.
Desired
degree of permanence estimates.
f.
When
is the estimated required.
g.
Part
- part of forecasting is desired, such as forecasting for the buyer group,
product group or geographical area.
2.
Development
Model
A
more modest presentation systems studied. In forecasting, the model is an
analytical framework which when inserted input data will generate an estimated
sales in the future (or what variables are considered. For example, if
companies want to predict sales of the "behavior" of her be linear,
the chosen model possible: sales = A + BX, where X represents the unit time, A
and B are the parameters that describe the position and the slope of the line
on the graph.
3.
Testing
Model
Models
are usually tested to determine the level of accuracy, validity and reliability
expected. This often includes historic application to the data, and preparation
of estimates for years now with real data available. The value of the model is
determined by the degree of accuracy of forecasting results with reality. In
other words, the test models intended to determine the validity or logic
predictive ability of the model.
4.
Application
Model
Analysis
of applying the model in this phase, historic data is included in the sales
model = A + BX, analysts apply techniques - mathematical techniques in order to
obtain A and B.
5.
Revisions
and Evaluations
Predictions
that have been made must be constantly revised and revisited. Repairs may be
necessary because of changes in the company or the environment, such as the
price level of enterprise products, the characteristics of the product,
advertising expenditures, the level of government expenditures, monetary policy
and technological advances. Evaluation on the other hand is a comparison
predictions with actual results to assess the provision or use of a methode
forecasting techniques. This step is necessary to maintain the quality of the
estimates of time when that will come.
2.1.7
Forecasting Techniques
The
techniques of forecasting by T.Hani handoko
(2000: 262):
1.
Qualitative
Techniques.
Qualitative
techniques are subjective or "judgmental" or based on estimates and
opinions of. Various sources of income for forecasting business conditions are
as follows:
a.
The
Executive
The
executives have the ability to provide useful input forecasting, especially of
the managers who have experience in the industry long enough or in similar
companies.
b.
People
Sales.
The
members of this group are still in touch with the customer, so it will be able
to estimate the purchasing plans, attitudes and their needs. The sales people
are also a source that can provide about tactics on a competitor now and
forecasts in the future.
c.
Clients
Subscriptions
(customers) that outputs (product or service) companies are sometimes willing
and eager to express their purchase plans. It is often found, especially for
companies that sell products to the market industry, and the information
provided the subscription is feedback for the company. Subscriptions may convey
this information personally to the executor and sales people, through letters
and phone charging a consumer survey questionnaires or private interviews.
d.
Etc
These
specialists are experts in various fields giving opinions are highly valued.
While a variety of qualitative forecasting
techniques that can be used briefly be described as follows:
a. Delphi method
Is
a technique that uses a systematic procedure to obtain a consensus of the
opinions of a group of experts. Delphi process is done by asking the members of
the group to give a series of predictions through their responses to a
questionnaire. Then the moderator collects and formulating list of new
questions and dealt to the group. So there is a process of "Learning"
for the group because they receive new information and no influence on the
pressure group or individual domination.
b. Market research
Forecasting
technique is useful especially when there is a shortage of data or historical
data is not reliabel. This technique is typically used to predict long-term
demand and sales of new products. Market research requires a series of steps as
follows:
-
Ensure
that the information sought.
-
Ensuring
information sources.
-
Establish
a method of procurement or collection of data, with personal interviews,
telephone surveys, mail surveys of observation, interviews and group panel, or
test the market.
-
Develop
and conduct a preliminary test measurement equipment.
-
Formulate
a sample.
-
Getting
information.
-
Perform
tabulation and analysis.
c. Analogy History
Forecasting
is done by using practice-historical experiences of a similar product.
Forecasting new products could be linked to the stages in the life cycle of
similar products.
d. Consensus Panel
The
idea was discussed by the group will produce better predictions than done by
someone. The discussion is conducted in meetings open exchange of ideas.
Participants may comprise executive, sales people, experts or subscription.
2.
Quantitative
Techniques
Quantitative
techniques are commonly used forecasting method to predict events in the future
by looking at the data in the past, this data series is a series of
observations of various variables according to the time which is usually
tabulated. Quantitative forecasting results is preferred because it provides a
view that is more real and more objective in the magnitude of the value of
forecasting. Various
quantitative forecasting techniques are as follows:
a. Freehand
With
this method the trend line is made freely without using a mathematical formula.
The curve trend "freehand" is depicted through data points and is the
easiest way of presenting. Forecast can be obtained simply by drawing a trend
line for the period of forecast. However, what seems to be adequate for a
company does not necessarily apply to another company, or this method has a
very high subjectivity so rarely used.
b. Least Squares (Least Squares)
The
method most often used to determine the equation of trend data because this
method produces what is mathematically described as a "line of best
fit". This trend line
has the properties:
-
The
sum of the whole vertical deviation of the data points to the line is zero.
-
The
sum of the squared deviations entire vertical historical data of the line is
the minimum.
-
The
line through the average X and Y.
The
formula in use:
For
linear equations, trend lines sought by the simultaneous settlement of the
value of a and b in normal following two equations:
Σ
Y = a + b n Σ X
Σ
XY = a + b X Σ Σ X2
When
the midpoint of the data as the basis years, then Σ X = 0 & retrievable
eliminated from equation above, so that it becomes:
Σ
Y
Σ
Y = n aa =n
Σ
XY
Σ
Σ X2 XY = b b =Σ X2
c. Moving Average
Forecasting
methods by combining data from several periods of the latest / last. The moving
average is obtained by summing and finding the average value of a specified
period, each time eliminating the value of the longest and add new value.
The
formula used is:
Information
:
Ft = Forecasting (Forecasting)
Ft = Forecasting (Forecasting)
X
= Data Period
n
= Duration
d. Weight Moving Average
A
method similar to the method of moving average, differing only in the addition
of weight to each data. The latest data are included in the calculation of the
average period given greater weight. The formula used is:
Information
:
A
= Weight of Largest
B
= Weight of Largest 2
C
= Weight of Largest 3, etc.
n
= Data Period
n-1
= Data 1st period before the last period
n
2 = Data of the 2nd period before the last period
e. Single Exponential Smoothing
A
method of moving average forecasts are weighing in on the past data with an
exponential manner. In this method of forecasting is done by other means
forecast last period plus the portion of the difference or error rate (denoted
by α) between the last period of real demand and forecast the most recent
period, the equation is:
Information
:
Ft
= Forecasting
Ft
- 1 = forecast for the previous period (t - 1)
α
= Smoothing constant (the portion of the difference)
At
- 1 = real inquiry earlier period.
2.2
Study Research Similar
This kind of study is taken from research that have
similar topics / variables that are and will be examined by the author.
1. Name: Angga Dwi Yuni Anto
NPM:
10207119
Title:
Forecasting Sales Starter Pack Xl Free On PT.Excelcomindo (Xl) Branch Kalimas
Bekasi
Supervisor:
Sri Kurniasih Agustin, SE., MM
Conclusion:
Based on sales forecasting XL card on PT.Exelcomindo Non Bekasi branch Kalimas
east for the month August 2010 the most precise method used is a method Weight
Moving Average, with great Weight Moving Average method of forecasting pedana
XL card for the month of August 2010 resulted in the value of forecasting
24,999 cards with a value of 188 card error while using Moving Average yield
value forecasting error value 189 24 849 with the card, and Exponential
Smoothing generate value forecasting error value 242 25 232 with the card.
2. Name: Eddie Wibowo
NPM:
10205387
Title:
Motorcycle Sales Forecasting Vario In PT.Tunggul Mitra Sejati
Bekasi
Supervisor:
Lies Handrijaningsih, SE, MM
Conclusion:
After forecasting conducted on the value of sales Motorcycle Vario In
PT.TunggulMitraSejati Bekasi Average, Weight Moving Average and Exponential
Smoothing. Forecasting using Exponential Smoothing is a method of the smallest
error or mistake by the error value of 7 units of value forecasting 141 units,
compared with the Moving Average method that generates an error value of 14
units of value forecasting 140 units and Weight Moving Average which generates
the error value of 14 units of the value of forecasting 141 units.
2.3
Analysis Tool
The analysis tool used is the Moving Average and
Moving Average Weight. Some of the assumptions of both of these methods are:
1. Data from consecutive time.
2. Using the data of the past.
3. Do not have equality.
4. Not suitable for data that are no
symptoms trend.
5. Can not keep up with changes
drastically.
1.
Moving
Average
Forecasting
methods by combining data from several periods of the latest / last. The goal
is to make the data into data fluctuates relatively stable so that the
fluctuation of the pattern or data to be smooth and relatively evenly (Hani
Handoko, 200: 157) Steps forecasting using Moving Average:
a.
Specifies
the number of periods to get the average price.
b.
Make
a calculation table
c.
Finding
the value of total mobile
d.
Finding
value in forecasting
The
formula used is:
Information :
Ft = Forecasting (Forecasting)
X = Data period
n = Timed
2.
Weight
Moving Average
A method similar to the method of
moving average, differing only in the addition of weight to each data. The
latest data are included in the calculation of the average period given greater
weight. The weakness of this method is the response can not be easily changed
without changing each of the weights.
The formula used is:
Information :
A = Weight of Largest
B = Weight of Largest 2
C = Weight of Largest 3, etc.
n = Data Period
n-1 = Data 1st period before the
last period
n 2 = Data of the 2nd period
before the last period
3.
Single
Exponential Smoothing (ES)
Exponential Smoothing is a meode
moving average forecasting that conduct declining exponential weighting of the
values of the older observations (Makridakis, 1993: 79).
In this method of forecasting is
done by other means forecast last period plus the portion of the difference or
error rate (denoted by α) between the last period of real demand and forecast
the most recent period, the equation is:
Information :
Ft = Forecasting
Ft - 1 = forecast for the
previous period (t - 1)
a = Smoothing constant (the portion
of the difference)
At - 1 = real inquiry earlier
period
4.
Forecasting
Errors
Forecasting error has two
elements that must be considered:
a.
The
difference between the real demand forecasting (error)
b.
Directions
mistake, that is, whether real demand are above or below forecast. There is a mistake commonly used
measure is the Mean Absolute Deviation (MAD), which is the size of finding the
difference between real and forecast demand with the average rate for a
predicted error is:
MAD = Error
N – n
Information :
N = Number of sales data
n = number of periods
CHAPTER
IV
DISCUSSION
4.1
Data and Research Object
4.1.1
A
Brief History of the Company
Simpatindo Multimedia is a subsidiary of Sarindo
Group established since October 29, 2002 with the legal form of a Limited
Liability Company. PT. Simpatindo Multimedia is a company engaged in trading
and distributing. At this time PT. Simpatindo Multimedia act as Authorized
Dealer Telkomsel. PT. Simpatindo MULTIMEDIA has carried on business that is
from the year 2002 - 2015 and has distributed four prime cards telkomsel
products and has a superior product that Simpati Sim Card.
Since 2002 PT. Simpatindo MULTIMEDIA already
distributes SIM cards Sympathy enough consistency, because the product was
distributed in accordance with its marketing channels. In doing marketing in
PT.SIMPATINDO MULTIMEDIA opened branches spread all over Indonesia with
marketing personnel experienced in the field. In this case the issuing
PT.TELKOMSEL Simpati Sim Card products that have been adjusted to the needs of
consumers and distributed directly by PT. MULTIMEDIA Simpatindo.
In this case there are four types of card products
that have been distributed by the prime PT.SIMPATINDO MULTIMEDIA, but only 1
products in featured, Simpati Sim card products. Because this product is
preferred by network facilities PT.TELKOMSEL by offering a strong signal and
spread throughout Indonesia. It can be seen from the number of products sold
more than other types of products. In other words that the Simpati Sim Card
products have a distinctive position for PT.SIMPATINDO MULTIMEDIA.
4.1.2
Organizational Structure
1. Director
a. As head of the company.
b. Formulating objectives and
determining overall company policy
c. Lead and oversee the development
of the company through the report - a report received and take the necessary
decisions.
d. Coordinate all the existing sections
in the company so as to create a harmonious cooperation and the achievement of
company objectives
e. Develop and establish plans,
objectives and strategies for the sale of short-term and long-term.
2. Financial Section
a. Accept and dispense cash for corporate
purposes
b. Making financial report
c. The company's financial control
3. Parts
Warehouse
Responsible
to the leading companies on the availability of goods.
4. Part
Sales
a. Responsible to corporate leaders
on sales transactions
b. Organize sales activities and promotions
in order to achieve maximum benefit.
c. Responsible for the delivery of
goods to the place of purchase orders.
d. Create sales reports.
5. The
Public Service
a. Section in charge of providing
services to consumers products company.
b. Give information in the service to
users of the company's products.
4.2
Discussion and Research
For each company forecasting have an important role
to market the products offered, precise sales forecasting whether or not to
become a reference for evaluating the targets that have been implemented by the
company.
At this writing, the author uses the method of
moving average, weight moving average and exponential methods. The data used by
sales 5 different SIM cards sympathy card with an internet mania prime
sympathy, sympathy starter pack talk mania, mania of smartphones sympathy card
is prime, prime cards sympathy sms mania, and starter pack sympathy nelpon
packages home. Here is a prime sympathy card sales data on PT. Multimedia
Simpatindo Cakung East Jakarta branch mill for 31 months from the date of 01
January 2013-15 July 2015.
Table
4.1
Starter
Pack of Simpati Sim Card Sales Data
The
period January 2013 - July 2015
Month
|
Sales
( Unit )
|
January
2013
February
March
April
May
June
July
August
September
October
November
December
|
972
988
990
1050
1010
1030
1080
1110
1125
1120
1300
1315
|
January
2014
February
March
April
May
June
July
August
September
October
November
December
|
1350
1415
1380
1400
1390
1420
1460
1500
1530
1600
1580
1620
|
January
2015
February
March
April
May
June
July
|
1615
1720
1744
1812
1822
1848
1860
|
Figure
4.2 Sales Chart PT.Simpatindo Multimedia
Year
2013
Figure
4.3 Sales Chart PT.Simpatindo Multimedia
Year
2014
Figure
4.4 Sales Chart PT.Simpatindo Multimedia
2015
Based on the above data, it can be seen selling Simpati
Sim Card PT.Simpatindo Multimedia whenever changes. To avoid instability in the
company should be able to predict exactly in sales in the coming year.
4.3
Calculation
of Sales Data
4.3.1
Sale
Forecasting PT. Simpatindo Multimedia Branch Office Penggilingan Jakarta Timur used
Metode Moving Average with 3 period.
Forecasting
is done to assist the planning and supervision of the prime Simpati card sales
, it helps the company's internal decision-making and is very useful for the
task of Top Managers . With the sales forecasting the company aware of the
possibility of activities in the future , so managers can seek redress in order
to efficient sales service . Application of forecasting with Moving Average Method
on PT.Simpatindo Multimedia Branch Penggilingan Cakung, East Jakarta Branch intended
to solicit sales forecasting Simpati Sim Card.
By using the method of Moving
Average 3 -month period , then the result was obtained by using the formula :
Description
:
Ft = Forecasting
X = Data Period
n = Time Period
Table 4.2
Calculation of
Sales Forecasting Simpati Sim Card August 2015 Method with Moving Average 3
period
Month
|
Sale
|
Forecasting
|
Error (e)
|
January 2013
|
972
|
-
|
-
|
February
|
988
|
-
|
-
|
March
|
990
|
-
|
-
|
April
|
1050
|
983,3
|
66,7
|
May
|
1010
|
1009,3
|
0,7
|
June
|
1030
|
1016,7
|
13,3
|
July
|
1080
|
1030
|
50
|
August
|
1110
|
1040
|
70
|
September
|
1125
|
1073,3
|
51,7
|
October
|
1120
|
1105
|
15
|
November
|
1300
|
1118,3
|
181,7
|
December
|
1315
|
1181,7
|
133,3
|
|
|
|
|
January2014
|
1350
|
1245
|
105
|
February
|
1415
|
1320
|
95
|
March
|
1380
|
1358,3
|
21,7
|
April
|
1400
|
1381,3
|
18,7
|
May
|
1390
|
1398,3
|
-8,3
|
June
|
1420
|
1390
|
30
|
July
|
1460
|
1403,3
|
56,7
|
August
|
1500
|
1423,3
|
76,7
|
September
|
1530
|
1460
|
70
|
October
|
1600
|
1496,7
|
103,3
|
November
|
1580
|
1543,3
|
36,7
|
December
|
1620
|
1570
|
50
|
|
|
|
|
January2015
|
1615
|
1600
|
15
|
February
|
1720
|
1605
|
115
|
March
|
1744
|
1651,7
|
92,3
|
April
|
1812
|
1693
|
119
|
May
|
1822
|
1758,7
|
63,3
|
Juny
|
1848
|
1792,7
|
55,3
|
July
|
1860
|
1827,3
|
62,8
|
August
|
-
|
1843,3
|
-
|
Total
|
1760,6
|
Calculations
Forecasting Monthly:
2013
1. April = 972 + 988 + 990 = 983,3
3
2. May = 988 + 990 + 1050 = 1009,33
3
3. June = 990 + 1050 + 1010 =
1016,7
3
4. July = 1050 + 1010 +
1030 =
1030
3
5. August = 1010 + 1030 + 1080 = 1040
3
6. September = 1030
+ 1080+ 1110 = 1073,3
3
7. October = 1080
+ 1110 + 1125 = 1105
3
8. November = 1110 + 1125+ 1120 = 1118,3
3
9. December = 1125 + 1120+ 1300 = 1181,7
3
2014
1. January = 1120 + 1300+ 1315 = 1245
3
2. February = 1300 + 1310+
1350 = 1320
3
3. March = 1310 + 1350+ 1415 = 1358,3
3
4. April = 1350 + 1415+ 1380 = 1381,3
3
5. May = 1415 + 1380+ 1400 = 1398,3
3
6. June = 1380 + 1400+ 1390 =
1390
3
7. July = 1400 + 1390+ 1420 = 1403,3
3
8. August = 1390 + 1420+ 1460 = 1423,3
3
9. September = 1420 + 1460+ 1500 = 1460
3
10. October = 1460 + 1500+ 1530 =
1496,7
3
11. November = 1500 + 1530+ 1600 = 1543,3
3
12. December = 1530 + 1600+ 1580 = 1570
3
2015
1. January = 1600
+ 1580+ 1620 =
1600
3
2. February = 1580
+ 1620+ 1615 = 1605
3
3. March = 1620 + 1615+ 1720 = 1651,7
3
4. April = 1615 + 1720+ 1744 =
1693
3
5. May = 1720 + 1744+ 1812 =
1758,7
3
6. June =1744 + 1812+ 1822 = 1792,7
3
7. July =1812 + 1822+ 1848 = 1827,3
3
8. August =1822+1848+1860 = 1843,3
3
MAD = ∑
Error
N
- n
=
1760,6
31 - 3
= 62,8
From
the above calculation using the method in the know Moving Average
sales forecasting results in August 2015 is as much as 1843 card and
forecasting error ( Mean Absolute Deviation ) was as much as 62.8 or 63 prime Simpati
Sim Card.
The deviation can be calculated
from the amount of the sales forecast range as follows :
The Range =
Ft – MAD < X < Ft + MAD
=
1843 – 63 < X < 1843 + 63
=
1780 < X < 1906
(Minimum Sales ) (MaximumSales)
From
the above calculation is unknown if the company will sell , it should be the
number of sales ranged from 1780 to 1906 Simpati Sim Card is prime .
4.3.2
Sales forecasting
Simpati Sim Card August 2015 at
PT . Simpatindo Multimedia Branch Milling East Jakarta . Method Using Moving
Average Weight By Weight 70 % and 30 %
To simplify the calculation in use
Weight Moving Average method used here considered that the two month period is
forecast was best achieved by using a weighting of 70% and 30 % .
With Weight Moving Average method with a weight of 70% and 30 % , then
the result was obtained by using the formula :
Description :
A = Bobot Terbesar
B = Bobot Terbesar Ke-2
n =
Data Period
n-1 = Data 1 period
before last period
n-2 = Data 2 period before last period
Table 4.3
Calculation of Sales Forecasting Psimapti Sim Card August
2015 With Weight Moving Average Method 2 periods .
Month
|
Sales
|
Forecasting
|
Error (e)
|
January2013
|
972
|
-
|
-
|
February
|
988
|
-
|
-
|
March
|
990
|
983,2
|
6,8
|
April
|
1050
|
989,4
|
60,6
|
May
|
1010
|
1032
|
-22
|
June
|
1030
|
1022
|
8
|
July
|
1080
|
1024
|
56
|
August
|
1110
|
1065
|
45
|
September
|
1125
|
1101
|
24
|
October
|
1120
|
1120,5
|
-0,5
|
November
|
1300
|
1121,5
|
178,5
|
December
|
1315
|
1246
|
69
|
|
|
|
|
January2014
|
1350
|
1310,5
|
39,5
|
February
|
1415
|
1339,5
|
75,5
|
March
|
1380
|
1395,5
|
-15,5
|
April
|
1400
|
1390,5
|
9,5
|
May
|
1390
|
1394
|
- 4
|
June
|
1420
|
1393
|
27
|
July
|
1460
|
1411
|
49
|
August
|
1500
|
1448
|
52
|
September
|
1530
|
1488
|
42
|
October
|
1600
|
1521
|
79
|
November
|
1580
|
1579
|
1
|
December
|
1620
|
1586
|
34
|
|
|
|
|
January2015
|
1615
|
1608
|
7
|
February
|
1720
|
1616,5
|
103,5
|
March
|
1744
|
1688,5
|
55,5
|
April
|
1812
|
1736,8
|
75,2
|
May
|
1822
|
1791,6
|
30,4
|
June
|
1848
|
1819
|
29
|
July
|
1860
|
1840,2
|
39,8
|
August
|
-
|
1856,4
|
-
|
Total
|
1155,1
|
Calculations
Forecasting Monthly
2013
1. March = ((0,7*988) +
(0,3*972)) =983,2
2. April = ((0,7*990) +
(0,3*988)) = 989,4
3. May = ((0,7*1050) +
(0,3*990)) = 1032
4. June = ((0,7*1010) +
(0,3*1050)) = 1022
5. July = ((0,7*1030) + (0,3*1010)) =1024
6. August = ((0,7*1080) + (0,3*1030)) =1065
7. September = ((0,7*1110)
+ (0,3*1080)) =1101
8. October = ((0,7*1125)
+ (0,3*1110) =1120,5
9. November = ((0,7*1120) + (0,3*1125)) = 1121,5
10. December = ((0,7*1300) + (0,3*1120)) = 1246
2014
1. January = ((0,7*1315) + (0,3*1300)) = 1310,5
2. February i = ((0,7*1350
+ (0,3*1315)) = 1339,5
3. March = ((0,7*1415) +
(0,3*1350)) = 1395,5
4. April = ((0,7*1380) +
(0,3*1415) ) = 1390,5
5. May = ((0,7*1400) +
(0,3*1380)) = 1394
6. June = ((0,7*1390) +
(0,3*1400)) = 1393
7. July = ((0,7*1420) +
(0,3*1390)) = 1411
8. August = ((0,7*1460)
+ (0,3*1420)) = 1448
9. September = ((0,7*1500)
+ (0,3*1460)) = 1488
10. October = ((0,7*1530) + (0,3*1500)) = 1521
11. November = ((0,7*1600)
+ (0,3*1530)) =1579
12. December = ((0,7*1580)
+ (0,3*1600)) =
1586
2015
1. January = ((0,7*1620)
+ (0,3*1580)) = 1608
2. February = ((0,7*1615) + (0,3*1620)) = 1616,5
3. March = ((0,7*1720) + (0,3*1615)) =1688,5
4. April = ((0,7*1744) +
(0,3*1720)) =1736,8
5. May = ((0,7*1812) +
(0,3*1744)) = 1791,6
6. June = ((0,7*1822) + (0,3*1812)) = 1819
7. July = ((0,7*1848) + (0,3*1822)) = 1840,2
8. August = ((0,7*1860) + (0,3*1848)) =
1856,4
MAD = ∑ Error
N - n
=
1155,1
31 - 2
= 39,8
From
the above calculation using the Weight Moving Average to know the results of
forecasting sales in August 2015 is as much as 1856 card and forecasting error
( Mean Absolute Deviation ) was as much as 39.8 or 40 prime Simpati Sim Card .
The deviation can
be calculated from the amount of the sales forecast range as follows :
The Range =
Ft – MAD < X < Ft + MAD
=
1856 – 40 < X < 1856+ 40
=
1816 < X < 1896
(Minimum
Sales )
(Maximum Sales)
From the above calculation is unknown if the company
will sell , it should be the number of sales ranging from 1816 up to 1896 Simpati
card is prime .
4.3.3
Sales
forecasting Prime Card Sympathy on PT . Multimedia Simpatindo East Jakarta
Branch Milling Using Methods Exponential Smoothing (ES) = 0,05
|
To simplify the calculation in use Method Exponential Smoothing .by using Method Exponential Smoothing, the result was obtained by using the formula :
Explanation :
Ft = Forecasting
Ft – 1 = The forecast For the Previos
Period ( t - 1)
Α = Smoothing Constant (Portion of different )
At – 1 = Real demand before period
Table
4.4
Calculation of Sales Forecasting Simpati Sim Card
August 2015 using the ES α = 0.05
Monthly
|
Sales
|
Forecast
|
Error (e)
|
January 2013
|
972
|
-
|
-
|
February
|
988
|
-
|
-
|
March
|
990
|
972,8
|
17,2
|
April
|
1050
|
973,7
|
76,3
|
May
|
1010
|
977,5
|
32,5
|
June
|
1030
|
979,2
|
50,8
|
July
|
1080
|
981,7
|
98,3
|
August
|
1110
|
986,6
|
123,4
|
September
|
1125
|
992,7
|
132,3
|
October
|
1120
|
999,3
|
120,7
|
November
|
1300
|
1005,3
|
294,7
|
December
|
1315
|
1020
|
295
|
|
|
|
|
January 2014
|
1350
|
1034,7
|
315,3
|
February
|
1415
|
1050,2
|
364,8
|
March
|
1380
|
1068,4
|
311,6
|
April
|
1400
|
1084
|
316
|
May
|
1390
|
1099,8
|
290,2
|
June
|
1420
|
1114,3
|
305,7
|
July
|
1460
|
1129,6
|
330,4
|
August
|
1500
|
1146,1
|
353,9
|
September
|
1530
|
1163,8
|
366,2
|
October
|
1600
|
1182,1
|
417,9
|
November
|
1580
|
1203
|
377
|
December
|
1620
|
1221,9
|
398,1
|
|
|
|
|
January 2015
|
1615
|
1241,8
|
373,2
|
February
|
1720
|
1260,5
|
459,5
|
March
|
1744
|
1283,5
|
460,5
|
April
|
1812
|
1306,5
|
505,5
|
May
|
1822
|
1331,8
|
490,2
|
June
|
1848
|
1356,3
|
491,7
|
July
|
1860
|
1380,9
|
291,7
|
August
|
-
|
1404,9
|
-
|
Calculations
|
8460,3
|
Calculations Forecasting Monthly :
2013
1.
March
= 972 + 0.05 (988 – 972 ) =972,8
2.
April
=
972,8 + 0.05 (990 – 972,8 ) =
973,7
3.
May =
973,7 + 0.05 (1050 – 973,7 ) = 977,5
4.
June =
977,5+ 0.05 (1010 – 977,5 ) =
979,2
5.
July = 979,2 + 0.05 (1030 – 979,2 ) =981,7
6.
August
= 981,7 + 0.05 (1080 – 981,7 ) =986,6
7.
September =
986,6 + 0.05 (1110– 986,6 ) =992,7
8.
October =
992,7 + 0.05 (1125– 992,7) =999,3
9.
November
= 999,3 + 0.05 (1120 – 999,3) =1005,3
10. December = 1005,3 + 0.05 (1300 – 10005,3 ) =
1020
2014
1.
January
= 1020+ 0.05 (1315– 1020) = 1034,7
2.
February =
1034,7+ 0.05 (1350 – 1034,7) =
1050 ,2
3.
March =
1050,2 + 0.05 (1415– 1050,2 ) =1068,4
4.
April =
1068,4+ 0.05 (1380– 1068,4) =1084
5.
May =
1084 + 0.05 (1400– 1084) =1099,8
6.
June =
1099,8 + 0.05 (1390 – 1099,8) =
1114,3
7.
July = (1114,3
+ 0.05 (1420 – 1114,3) = 1129,6
8.
August =
1129,6 + 0.05 (1460 – 1129,6) =
1146,1
9.
September =
1146,1+ 0.05 (1500 – 1146,1) =1163.8
10. October = 1163,8
+ 0.05 (1530 – 1163.8) =1182,1
11. November =
1182,1 + 0.05 (1600– 1182,1) =1203
12. December = 1203
+ 0.05 (1580 – 1203) =
1221,9
2015
1.
January =
1221,9 + 0.05 (1620 – 1221,9) =1241,8
2.
February = 1241,8 + 0.05 (1615 – 1241,8) = 1260,5
3.
March =
1260,5 + 0.05 (1720 – 1260,5) =
1283,5
4.
April
=
1283,5 + 0.05 (1744– 1283,5) =1306,5
5.
May =
1306,5 + 0.05 (1812– 1306,5) =1331,8
6.
June = 1331,8 + 0.05 (1822– 1331,8) = 1356,3
7.
July = 1356,3 + 0.05 (1848 – 1356,3) = 1380,9
8.
August=
1380,9 + 0.05 (1860 – 1380,9) =
1404,9
MAD = ∑
Error
N
- n
=
8460,3
31 - 2
=
291,7
From
the above calculation using the Exponential Smoothing to know the results of
forecasting sales in August 2015 was as much as 1404.9 or 1405 card and
forecasting error ( Mean Absolute Deviation ) was as much as 291.7 or 292 prime
Sympathy cards .
The deviation can be calculated from the range of the amount of the sales forecast as follows :
Forecasting = Ft – MAD < X < Ft
+ MAD
= 1405 – 292 < X < 1405+ 292
= 1113 <
X < 1697
(Minimum Sales) (Maksimum Sales)
From the above calculation is unknown if the company will sell , it should be the number of sales ranged from 1113 until 1697 SIM Card Sympathy.
4.4
Summary
Calculation Results
Table 4.5
Summary Calculation Results
Forecasting Method
|
Moving Average
|
Weight Moving Average
|
Eksponential
Smoothing
|
Sales
( August 2015)
|
1843
|
1856
|
1405
|
Mean Absolut Deviation
|
63
|
40
|
292
|
sales forecast
|
1780< X <1906
|
1816< X<1896
|
1113< X <1697
|
Explanation :
Moving Average method, forecasting sales that may occur in August 2015 as many as 1843 cards and Mean Absolute Deviation much as 63 prime Sympathy cards. While the range of sales of 1780sampai dengan1906 card. If the range of the sale of the cards in 1780 under prime Simpati Sim Card sales said to be bad, and if the sale of Simpati Sim Card above 1906 then the sale is said to be good.
Weight Moving Average Method method, forecasting sales that may occur in August 2015 as many as 1856 cards and Mean Absolute Deviation 40 prime Simpati Sim Card. While the range of sales of 1816 until 1896 card. If the range of the sale of the cards in 1816 under prime Sympathy card sales said to be bad, and if the sale of SIM cards Sympathy cards above 1896 then the sale is said to be good.
Methods methods Exponential Smoothing, forecasting sales that may occur in August 2015 as many as 1405 cards and Mean Absolute Deviation 292 prime Sympathy cards. While the range of sales of 1113sampai dengan1697 card. If the range of the sale under the sale of starter packs 1113kartu Sympathy is deplorable, and if the sale of SIM cards Sympathy cards above 1697 then the sale is said to be good.
So we can conclude that the method closest to the truth or the smallest error rate is the Moving Average Weight. Weight Moving Average method is better than the method Moving Average and Exponential Smoothing Methods. So if the forecast sales of the Method of Weight Moving Average forecasting error is relatively small so the results of this forecasting method will guarantee its accuracy. Methods Weight Moving Average forecast sales of cards for the month of August 2015 as many as 1856 cards and Mean Absolute Deviation 40 prime Sympathy cards. While the range of sales of 1816sampai dengan1896 card. If the range of the sale of the cards in 1816 under prime Sympathy card sales said to be bad, and if the sale of SIM cards Sympathy cards above 1896 then the sale is said to be good.
From the results that have been obtained to minimize the risk of loss / excess demand, then the company should be able to sell at least 1816 and maximum prime cards Sympathy Sympathy 1896 SIM Card. It can be seen from the sales range Weight Moving Average Method.
Of the three methods used to perform sales forecasting acquired method is the best method Moving Average Weight. Because the error rate is smaller than the Moving Average Method and Exponential Smoothing Method.
CHAPTER
V
CONCLUSION
CONCLUSION
5.1 Conclusion
Based on the results of the Third calculation
method, the authors draw the following conclusion:
The method closest to the truth or the smallest
error rate is the Moving Average Weight. Weight Moving Average method is better
than the method Moving Average and Exponential Smoothing Methods. So, if the
forecast sales of the method of moving average error in forecasting relatife
smaller, so the results of this forecasting method will guarantee its accuracy.
Methods Weight Moving Average forecast sales of cards for the month of August
2015 as many as 1856 cards and Mean Absolute Deviation 40 prime Simpati Sim
Cards. While the range of total sales in 1816 up to 1896 cards. If the range of
the sale of the cards in 1816 under prime Simpati Sim Card sales said to be
bad, and if the sale of Simpati Sim cards above 1896 then the sale is said to
be good. From the results that have been obtained to minimize the risk of loss
/ excess demand, then the company should be able to sell at least 1816 and
maximum 1896 SIM Card. It can be seen from the sales range Weight Moving Average
Method.Of the three methods used to perform sales forecasting acquired method
is the best method Moving Average Weight. Because the error rate is smaller
than the Moving Average Method and Exponential Smoothing Methods.
5.2 Suggestions
After researching, calculating, analyzing and
conclusions of forecasting sales at PT.Simpatindo Multimedia Branch
Penggilingan East Jakarta, the author tries to give advice, if someday the
company wants to do sales forecasting, you should use Method for Weight Moving
Average error rate is smaller than method of Moving Average and Exponential
Smoothing Methods. Although the results are not precise forecasting up to 100%
but can be considered to make the planning and the target company's activities
further in increasing sales strategy.