Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Qiurui Liu2
Intelligent Retail Forecasting System for New
Clothing Products Considering Stock-out
DOI: 10.5604/12303666.1227876
1
College of Fashion and Design, Abstract
Donghua University, Improving the accuracy of forecasting is crucial but complex in the clothing industry, espe-
Shanghai, China cially for new products, with the lack of historical data and a wide range of factors affecting
E-mail: huanghedhu@gmail.com; demand. Previous studies more concentrate on sales forecasting rather than demand foreca-
2
College of Public Administration, sting, and the variables affecting demand remained to be optimized. In this study, a two-stage
Huazhong Agricultural University, intelligent retail forecasting system is designed for new clothing products. In the first stage,
Wuhan, China demand is estimated with original sales data considering stock-out. The adaptive neuro fuzzy
E-mail: qiurui.liu@hotmail.com inference system (ANFIS) is introduced into the second stage to forecast demand. Meanwhile
a data selection process is presented due to the limited data of new products. The empirical
3
Faculty of Management, data are from a Canadian fast-fashion company. The results reveal the relationship between
McGill University, demand and sales, demonstrate the necessity of integrating the demand estimation process
Montreal, Canada into a forecasting system, and show that the ANFIS-based forecasting system outperforms
the traditional ANN technique.
GENERAL PROBLEMS OF THE FIBRE AND TEXTILE INDUSTRIES
Key words: intelligent forecasting system, demand estimation, stock out, adaptive neuro
fuzzy inference system, new clothing product.
motion, macroeconomics, etc. [3]; (4) models/techniques have been created for
acquiring an accurate demand pattern is the clothing industry.
not easy because of stock-out, thus using
historical sales data instead of demand Traditional techniques
data to implement forecasting is a com- An inspection of previous studies reveals
mon way. that classical statistical methods are ex-
tensively used [5-7]. However, most
To explore the issue of clothing retail methods such as regression models are
forecasting, some papers have investigat- only efficient for seasonal or cyclical
ed statistical methods, while others have data [8], and limited to a linear struc-
studied artificial intelligence (AI) tech- ture. Moreover the demand for clothing
niques. In recent years, hybrid techniques products is affected by numerous factors,
have aroused the interest of experts. so that statistical methods may not be
Among the various hybrid techniques, capable of achieving a satisfactory re-
the adaptive neuro fuzzy inference sys- sult. Moreover a substantial amount of
tem (ANFIS) is a competent alternative data are often required to generate good
that combines the advantages of the ar- results, but the data available for new
tificial neural network (ANN) and fuzzy clothing products are limited. Therefore
system [4]. In addition, previous studies AI techniques have gained more atten-
heavily focus on developing forecasting tion [9-11]. Despite the fact that AI tech-
models/techniques, but ignore dataset niques could obtain satisfactory results
Introduction processing. In fact, the purpose of retail in many studies, it is not easy to achieve
forecasting is to satisfy consumer de- a balance between result accuracy, high
For the clothing industry, it is critical to
mand. Therefore integrating the demand computational speed and system sta-
improve the accuracy of demand fore-
estimation process into the forecasting bility. Consequently to combine the ad-
casting [1]. A good forecasting system
system should be taken into account. vantages of different techniques, hybrid
could avoid problems such as lost sales,
inadequate inventory level, inferior cus- techniques were investigated.
As a result, this study aims to design an
tomer service, etc. However, four main
intelligent system for the demand fore- Hybrid techniques
factors increase the difficulty in cloth-
casting of new clothing products. This
ing forecasting: (1) The clothing supply A great quantity of studies have presented
system comprises of a demand estima-
chain is considered inflexible and com- that hybrid techniques lead to better per-
tion stage and demand forecasting stage.
plex with many sectors and companies; formance. As a hybrid technique, ANFIS
The empirical data are from a Canadian
[2] it results in the high sensitivity of combines the knowledge representation
fast-fashion company.
the bullwhip effect; (2) a large number ability of the fuzzy system and the learn-
of new products introduced to each col- ing ability of ANN. Consequently ANFIS
lection leads to extensive historical data has been widely used for forecasting in
Forecasting techniques
for past products, but little data for new different industries [12-14]. However,
products; (3) The sales data of clothing
in the clothing industry
there is not an adequate quantity of stud-
products is usually noisy and affected by As time goes on and computer technolo- ies in the literature on retail forecasting
multiple factors, such as weather, pro- gy develops, more and more forecasting with ANFIS in the clothing industry.
10 Huang H, Liu Q. An Intelligent Retail Forecasting System for New Clothing Products Considering Stock-out.
FIBRES & TEXTILES in Eastern Europe 2017; 25, 1(121): 10-16. DOI: 10.5604/12303666.1227876
Method and dataset The data selection process is explained it indicates that stock-out may happen.
as follows: We also realize that the demand on each
Proposed system and data selection n Based on the concept of ANFIS and day of the week varies a lot. For instance,
As an important information flow, de- data splitting theory, two product sales at the weekend are usually stronger
mand information greatly influences the groups are identified: group 1 has 56 than those on weekdays. Then each day
performance of clothing retail compa- items with full-year data from 2014, of the week has different sales weight.
nies. However, unlike other areas where and group 2 has 8 items with data We express the weekly demand of each
the past is a great predictor of the future, from several months in 2015. item D as below:
m
k
such as grocery or hardware retail, there n All products belong to the same prod-
D = ∑ D (1)
m T m
is no amount of historical data or experi- uct department, and there are three k t =1 tk
ences that can truly and effectively pre- product classes in group 1. 7 7
al techniques have certain disadvantages of comparable items, such as ZARA week k. m = (1,2,…,M); t = (1,2,…,T);
such as an inability to address a complex [15]. k = (1,2,…,K).
non-linear structure, a need for more his- n For each item, daily data from each D – the demand of item m in week k
m
torical data, the production of relatively store are collected rather than aggre- which aggregates the demand of all
worse results, etc. In this section, a two- gated data. stores.
stage intelligent retail forecasting system S tkd – sales quantity of item m in store t
m
that could remedy the shortcomings of In addition, the life cycle of fast-fashion on day d of week k.
V tkd – the binary variable equals 0 when
m
traditional techniques is proposed, see products is short, usually 6-10 weeks
Figure 1. In the first stage, demand is es- [16]. In our database, the products in stock-out may exist (daily inventory
timated with original sales data consider- group 2 include a complete product life = 0), otherwise 1 (daily inventory > 0).
ing stock-out. In the demand forecasting cycle of 10 weeks. Therefore, in this W d – salces weight of day d.
stage, ANFIS is employed and the sig- study, the forecasting time period is 10 d – index of the day, d = (1,2,…,7),
nificant variables affecting demand are weeks with a weekly horizon. Monday is 1 and Sunday – 7.
selected as the ANFIS inputs. Given the S’d – aggregated sales quantity on day d,
data of new clothing products is limited, Demand estimation used to compute Wd.
a data selection process is integrated in The occurrence of stock-out results in in-
this system. equality between sales and demand [17]. The basic idea behind this approach is
In other words, if there is no stock-out, consistent with one of the methods dis-
To validate the effectiveness of this sales quantity can be regarded as de- cussed in [18], although our approach is
system, we applied it to a Canadian mand quantity. Therefore the basic idea slightly different. To generate the sales
fast-fashion company. The company has of stage one is using the sales data with- weights, the daily sales of group 1 are
hundreds of stores across North Amer- out stock-out to deduce demand. In our aggregated by each day of the week, then
ica and a large Asian sourcing base. database, if the daily inventory is zero, the aggregated sales are divided by the
Demand
Original Demand
Estimation
Sales Data Data
Model
Actual
Data Selection
Value
Variable 1
ANFIS
Variable N
Parameters adjustment
weekly
ggregated sales are divided by the total sales of the whole weekly demand
demand
week, see obtained
obtained
Equation instead of
3. instead of weekly
weeklysales
saleswas
wasused
usedasasthe
theinputs
inputsofofthe
thesecon
seco
23.51% 20%
20%
19.66%
Sales weight
15%
15%
Sales weight
15.33%
10%
10% 12.44%
10.14% 10.65% 5%
8.27%
5%
0%
MON. TUES. WED. THURS. FRI. SAT. SUN.
0% Day
MON. TUES. WED. THURS. FRI. SAT. SUN.
Day Class A Class B Class C
Fig.
Figure 2. Sales 2 Salesfor
weights weights for each
each day of theday of the week
week. Fig.33Sales
Figure 3. Sales Fig. Sales
weights weightsfor
for weights
different for differentclasses
different
classes. classes
The three days with the highest sales are Saturday, Friday
3.3.and
3.3. Thursday,
Demand
Demand with
forecasting
forecasting
total sales of the whole week, see Equa- Equation 1. The weekly demand obtained ployed, which has become common prac-
ales weights of 23.51%, 19.66%, and 15.33%, respectively. Surprisingly
ItIt is
is sales
well sales on
well was
known
known that AIin-
techniques, suchas
asimplementation
thefuzzy
fuzzysystem
system and ANN,have
havetht
tion 3. Figure 2 shows the results. instead of weekly usedthat AI
as the techniques,
tice in such
ANFIS the and
[12]. ANN,
All
puts of the second stage. functions were carried out in MATLAB.
Sunday are relatively
The three low.
days The
with main reasonsales
the highest might
are be that retail stores of
capability
capability have
of shorterhuman
imitating
imitating human reasoning.
reasoning.ANN
ANNhas
hasaastrong
strongcapability
capabilityofoflearning
learningwit
w
Saturday, Friday and Thursday, with sales Demand forecasting The structure of a typical ANFIS is
usiness hoursweights of 23.51%,
on Sunday 19.66%,
in North and 15.33%,
America. It is well
Then another known
question that
is
parallel data. AI techniques,
whether
data. However, the
However, given such shown in is
given knowledge
knowledge Figure 4. For into
is integrated
integrated ease of
into thetheillustra-
wholenetwork
networkana
parallel whole
respectively. Surprisingly sales on Sunday as the fuzzy system and ANN, have the tion, it is assumed that each variable is
are relatively low. The main reason might capability of imitating human reasoning. fuzzed by two fuzzy sets. The node func-
ales weights observed would vary depending on the products cannotin different
be broken
broken classes,
up into
intolearn-
individual parts; the network isis essentially
essentially
be that retail stores have shorter business ANN has cannot be
a strong capabilityup of individual tionsparts;
of eachthe network
layer are explained below: aablack-box
black-bo
hours on Sunday in North America. Then ing with parallel data. However, given
ven if they belong
anothertoquestion
the sameis department.
whether theWe then examined
sales knowledgeThethe is sales weights
integrated
fuzzy system intoisfor
the
good whole
at reasoning
reasoning with the linguistic information i obtainedfrom
fro
fuzzy system is good at ANFISwithinputs thearelinguistic
represented as neuron xobtained
information
weights observed would vary depending network and cannot be broken up into in- (i = 1, 2, …, n). Here, they are the factors
on the
he three classes products
of group 1. in
Asdifferent
expected,classes, even quantity
the sales dividual parts;class
ofexpert
each the network is essentially affecting demand. Each neuron simply
is different,
expert knowledge.
knowledge. However, the
However, the learning
learningcapability
capabilityisisinferior inferiorandandcannot
cannotadapt
adaptitse
its
if they belong to the same department. a black-box. The fuzzy system is good at passes external crisp signals to Layer 1.
ut the sales We then examined
weights the salesconsistent,
are remarkable weights forsee Figure
reasoning 3. with the linguistic that
It demonstrates information
the three classes of group 1. As expected, obtained to from
a newnew environment
environment
expert knowledge.[19]. [19].
How- As aaLayer
As hybrid
hybrid technique, first
technique,
1 (Fuzzification first proposed
proposed
Layer): Neuronby Fby [20],ANFI
[20], ANF
ij
the sales quantity of each class is differ- ever, the learning capability is inferior (i = 1, 2, …, n; j = 1, 2) is the linguistic
roduct class ent,
has but
no significant impact
the sales weights areonremarkable
sales weights,
andand hence
cannot theitself
adapt salestoweights
a new environ-
combines
combines the
the learning
learning ability
ability of oflabel
ANN
ANN corresponding
with
with the to
the semanticthe fuzzytransparency
semantic set asso-
transparency ofof fuzz
fuz
consistent, see Figure 3. It demonstrates ment [19]. As a hybrid technique, first ciated with the input. The neuron func-
n Figure 2 can
thatbeproduct
used for each
class hasitem. After obtaining
no significant the salesby
im- proposed weights, the weekly
[20], ANFIS combines the tion can be modelled by the fuzzy mem-
systems. Meanwhile
Meanwhile the the domain
domain knowledge
knowledge represented
representedasasfuzzy fuzzyrulesrulesandandlinguisti
pact on sales weights, and hence the sales learningsystems. ability of ANN with the seman- bership function (MF).
linguis
emand of eachweights
item ininFigure 2 canwas
each store be used for each
computed tic transparency
by Equation 2, and thenof fuzzy systems. Mean-
the weekly
item. After obtaining the sales weights, while the variables
variables is
is integrated
domain knowledgeintegrated into
into ANFIS.
represented ANFIS. In In this
this1ijstudy,
Output = µ Fij (the
study, i ) =Sugeno-type
xthe membership fuzzy
Sugeno-type fuzzyinferenc
inferen
the weekly demand of each item in each as fuzzy rules and linguistic variables is function format, i = 1, 2, … , n, j = 1, 2.
emand of eachstoreitem
was that aggregates
computed all stores
by Equation was generated
2, and integrated by Equation
into ANFIS. In1.thisThe study,thethe Mamdani-type
system
system (FIS)
(FIS) rather
rather than
than the Mamdani-type FIS FIS isis employed,
employed, which which has has becom
becom
then the weekly demand of each item that Sugeno-type fuzzy inference system (FIS) Layer 2 (Rule Layer): Each neuron wl
aggregates all stores was generated by rather than the Mamdani-type FIS 7 is em- (l = 1, 2, …… , L) corresponds to a fuzzy
common
common practice
practice in
in ANFIS
ANFIS implementation
implementation [12].
[12]. All
All functions
functionswere
werecarried
carriedout
outi
MATLAB.
MATLAB.
Layer 1 Layer 2 Layer 3 Layer 4 Layer 5
F 11 w1 x1 xn
w1
x1 F 12 Π N y1
w1 * y1
• • • • •
Input • • • • • ∑ y Output
• • • • •
xn F n1 Π N yL wL * yL
wL
F n2 wL x1 xn
Promotion Level
Demand triMF Constant
Forecasting f(u)
Size Preference (sugeno)
gaussMF Linear
Allocation Status Demand Quantity
gbellMF
Calendar Factor
rule and functionally computes the firing but also determines the target custom- In the first step, determining the train-
strength. The neuron function should be ers. It has three fuzzy sets: high, medi- ing and validation datasets is important.
a certain T-norm function, such as, um and low. Given comparable products have similar
n Promotion Level: Some previous demand patterns, it is reasonable to use
n
studies used “promotion” and “no such data rather than random product
Output 2 l = wl = ∏ µ F i ( xi ) ,
l
i =1 promotion” as the fuzzy sets. In fact, data for forecasting. Consequently group
l = 1, 2, … , L. for a given product, it sets the final 1 data were used as training data to build
sales price after more than one mark- the networks. Group 2 data were used as
Layer 3 (Normalization Layer): Neuron down, and the promotion level varies validation data to validate the reliability
wl (l = 1, 2, … , L) determines the nor- between normal weekends and big of the networks. Furthermore given the
malised firing strength. For a given rule, holidays. Meanwhile different promo- variables were fuzzed by several fuzzy
the ratio between the firing strength and tional levels have different impacts on sets, with the help of company experts,
the sum of the firing strengths of all rules customer consumption. Therefore the numerical scales were used to convert
represents the normalised firing strength. “promotion level” was introduced, linguistic labels to numerical data.
L which is associated with three fuzzy
Output 3l = wl = wl ∑ w, l = 1, 2, … , L.
l sets: high, medium and low. In the second step, a Sugeno-type FIS
l =1 n Size Preference: Unlike colour pref- (see Figure 5) is constructed for ANFIS.
Layer 4 (Defuzzification Layer): Neurons erence, which changes every year or Determining the MF type of input vari-
yl (l = 1, 2, … , L) compute the conse- even every season, size preference is ables is an important but complex task.
quent of the rules. The neuron function in relatively stable. The main reason is To obtain the best performance, differ-
this layer is a linear function. that the target customers in a specific ent experiments were prepared for four
city are stable. Three fuzzy sets were commonly used MFs: the triangular MF
Output 4 l = wy l l l
l l = wl ( p 0 + p 1 x1 +…+ p nxn ) determined: high, medium and low. (triMF), trapezoidal MF (trapMF), gen-
l = 1, 2, … , L. n Allocation Status: In the planning eralised bell MF (gbellMF) and Gaussi-
stage, not every item is allocated to an MF (gaussMF). The output function
Layer 5 (Summation Neuron Layer): each store. Some items are allocated has two types, constant and linear, and
Neuron y determines the final output of to all stores, while others are not. The experiments prepared for both functions
ANFIS by computing the sum of the re- allocation strategy also affects the de- were conducted as well. The 8 different
sults of all defuzzification neurons. mand pattern. Two fuzzy sets were MF combination experiments used to
L L L
identified: normal and special. identify the best forecasting capabili-
Output 5 = y = ∑ wl ∗ yl = ∑ wl ∗ yl ∑w l n Calendar Factor: It is easy to observe ty of the ANFIS structure are shown in
l =1 l =1 l =1 that holidays or big events influence Figure 6.
As discussed earlier, neuron xi are the demand. For example, the sales on
inputs of ANFIS, which are the factors Black Friday are extremely high, In the training step, the ANFIS learning
affecting demand. Therefore selecting whereas those on normal workdays algorithm is used to optimise FIS param-
the appropriate factor is important. Fur- are low. This variable has two fuzzy eters. The hybrid algorithm, consisting
thermore it is usually difficult to analyse sets: normal and special. of the least squares approach and back-
qualitative variables with conventional propagation gradient descent approach,
methods. However, ANFIS is a powerful It should be noted that although colour was utilised. The ANFIS structure with
technique to cope with both quantitative preference also has a great influence on 5 inputs and 1 output was trained for
and qualitative variables. By interview- demand, the large number of existing 100 epochs with a 0.01 error tolerance.
ing company experts and reviewing pre- colours and new ones introduced every Considering the overfitting problem, the
vious literature, five factors were deter- season make it impractical to identify the model was run ten times for each MF
mined. fuzzy sets. After determining the five var- combination with varying the parame-
n Original Price Level: This variable iables, four main steps were implemented ter. Group 1 data were used for this pro-
not only affects the demand pattern, to build a robust ANFIS. cess. With different experiments, Gauss
planning, allocation
Table 3. Forecasting performancesdecision, inventory management, etc.
of ANFIS and ANN. If comparison
Results products are always
For most clothing companies, past sales
MAPE MAPE MSE MSE data are usually used for product plan-
Product No. Difference Difference
available
1
for sales29.05%
(ANFIS)
7.45%
and without
(ANN)
21.60%
stock-out,
(ANFIS)
4.450.55
then the
(ANN)
35,608.07 31,157.52
demand
ning, and salesdecision,
allocation information
inventory
management, etc. If products are always
2 9.02% 31.70% 22.68% 6,098.67 38,900.83 32,802.16 available for sales and without stock-out,
3 17.27% 37.80% 20.53% 27,468.70 84,127.40 56,658.70 then the demand and sales information
would 4 be identical.
6.37% However,
28.42% 22.05%due 3,291.09
to the demand
32,169.09 for clothing
28,878.00 wouldproducts
be identical. being affected
However, due to
5 10.72% 35.69% 24.97% 6,765.71 76,841.22 70,075.51 the demand for clothing products being
6 8.59% 33.41% 24.82% 5,642.73 58,466.45 52,823.72 affected by multiple factors and many
by multiple
7 factors40.88%
11.42% and many 29.46%factors are 104,590.77
7,654.59 unpredictable96,936.18 or factors
uncontrollable,
are unpredictable orstock-out
uncontrolla-
8 14.43% 32.90% 18.47% 9,851.00 42,346.01 32,495.01 ble, stock-out is rarely avoided, causing
Average 10.66% 33.73% 23.07% 8,902.88 59,131.23 50,228.35 the amount sold to be less than the ac-
is rarely avoided, causing
Data source: calculated using MATLAB the amount sold to be less than
tualthe
demand actual demand
[17]. Therefore one [17].
of the
main purposes in this study is to consider
stock-out when implement forecasting.
MF and Constant MF were chosen for
Therefore one of the main purposes
inputs and outputs, respectively. Since
Results and this study is toThe
in discussion first stage is estimating demand data
consider stock-out when
based on original sales. Aggregated de-
each rule in ANFIS is represented as “If Technique comparison mand and sales data of the training and
input, Then output”, 108 Sugeno–type The traditional ANN technique was in- validation datasets are shown in Fig-
implement forecasting.
FIS rules were generated. The first
Table 1 shows troducedstage is estimating
to facilitate a comparison withdemandure 7, data based
and Table ondata
2 presents original
of a spe-
example rules. the system proposed. To construct the cific product.
ANN network, the forecasted demand
sales. Aggregated
The last step is measuring thedemand
forecasting and sales
value data byofthethe
is generated training
vectors of the and validation
As expected, the demanddatasets
quantitiesare
are
capacity of the system. Group 2 data were five variables (original price level, pro- significantly higher than those for sales,
used, which include 8 items with a com- motion level, size preference, allocation which means that the sales data cannot
shown in life
plete product Figure
cycle of7,
10and
weeks.Table 2 presents
status datafactor),
and calendar of aand specific
the out- product.
accurately reflect real customer demand.
50,000 7,000
Quantity (unit)
Quantity (unit)
6,000
40,000
5,000
30,000 4,000
3,000
20,000
2,000
10,000
1,000
0 0
W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10
Week Week
Sales Demand Sales Demand
technique.
Product 3 Product 8
1,800 1,200
1,600
1,000
1,400
Quantity (unit)
1,200 800
Demand
1,000
600
800
600 400
400
200
200
0 0
W1 W2 W3 W4 W5 W6 W7 W8 W9 W10 W1 W2 W3 W4 W5 W6 W7 W8 W9 W10
Week Week
Actual ANFIS ANN Actual ANFIS ANN
Furthermore
In other the just
words, if the company forecasting
uses different capability wasbutalso
impacts on demand, compared
they are with the
rate and unreliable, company’s
and this task would
sales data as the input of its forecasting too numerous to establish an exhaustive be extremely tedious if the number of
system, it would lead to a certain amount list, and some are consistently unavail- new products is large. More importantly,
current approach.
of lost sales. Eventually, theCommercial
total revenue able, forecasting
such as weather,software is integrated
political factors the companyinto
failedtheir enterprise
to consider the issue
must be affected. As a result, it is essen- and competitors’ strategies [21]. Nev- of demand estimation, and thus their re-
tial to integrate the demand estimation ertheless, apart from the special points, sults are only sales quantities. With such
resource
stage into the planning (ERP) system.
forecasting system. However,
the ANFIS-based it generally
system can is well a consideration,
known that it is notcommercial
appropriate to
achieve values that are close to the real compare the demand quantities forecast-
In addition, forecasting performanc- demand. The comparison results clearly ed by our system and the sales quantities
es were compared between the AN- demonstrate that the system proposed the company computed by using MSE
software
FIS-based systemis usually designed
proposed (ANFIS, for universal
for outperforms companies
the traditional ANN tech- and is notHowever,
and MAPE. a good fitbetween
the gap for a
simplification)
curves strongly indicatesand ANN, see Table
the lost sales3.ignored nique.
by the company, see Figure 9. Such a curves strongly indicates the lost
the two
sales ignored by the company, see Fig-
specific
According to company.
the MAPE andMoreover
MSE values, this software
Furthermore is inadequate
the forecasting capability for
ure new
9. Suchproduct
a situation forecasting,
would result in
situation would result in system
the ANFIS-based enormous losses inwasthe
significantly alsolong run. with
compared In fact, some clothing
the company’s enormous losses in the long run. In fact,
increases the forecasting accuracy com- current approach. Commercial forecast- some clothing companies have attempted
pared with ANN. However, we noticed ing software is integrated into their en- to develop their individual customised
companiesthushave
that
company
attempted
sometimes
experts
the to develop
forecasting
needtheirto
results
make adjustments
individual customised
terprise resource
with
forecasting
planning (ERP)
theirforecasting
subjective
system. system
knowledge to a
system [21] based on the
are not very close to the real demand, see However, it is well known that com- company’s actual situation with the con-
Figure 8. The forecasted values of prod- mercial software is usually designed for sideration of more factors. If the demand
[21] based on the company’s
baseline computed actual
bysituation with the The
thefarsoftware. consideration
results of more factors.inaccurate
If
uct 3 in weeks 2 and 4 are relatively universal companies and is are usually
not a good estimation process isand unreliable,
integrated into their
from the real demand compared with the fit for a specific company. Moreover this customised system, we believe lost sales
the demandperformance
estimation process
of other weeks.isSuch
integrated
a situ- into theiris customised
software inadequate forsystem, we believe
new product and end-of-season inventories could be
and
ationthis
is alsotask would
observed be extremely
with product tedious
8 in forecasting, if the number
thus company experts needof new
reduced,products
meanwhile is the large. More
retail efficiency
week 7. A possible reason is that the var- to make adjustments with their subjective and total revenue could be improved.
lost sales iables
and may end-of-season inventories
not fully encapsulate demand could be reduced,
knowledge to a baselinemeanwhile
computed by the the retail
behaviour. Indeed different factors have software. The results are usually inaccu-
importantly, the company failed to consider the issue of demand estimation, and thus
Conclusions
efficiency and total revenue could be improved. To sum up, by first observing the sales
and demand patterns of datasets, the
their results 12,000 are only sales quantities. With such a consideration, it is not appropriate
system proposed could significantly re-
10,000 duce lost sales. In addition, the forecast-
ing capabilities of the system proposed
to compare 8,000
the demand quantities forecasted by our system and were
and ANN the compared.
sales quantities
The results
Quantity (unit)
6,000
demonstrate that the system proposed
outperforms ANN. Moreover the fore-
casting performance of the system pro-
the company 4,000
computed by using MSE and MAPE. However, posedthe
was gap between
compared the two
with the company’s
2,000
current approach. Given the company’s16
0 approach forecasted only with respect to
P1 P2 P3 P4 P5 P6 P7 P8 sales, lost sales were observed from the
Product gap between the demand quantities our
system forecasted and the sales quanti-
Intelligent System (Demand) Company Approach (Sales) ties the company computed. Neverthe-
less more data is required to obtain robust
Fig. 9.9 Forecasting
Figure Forecastingresultsresults of intelligent
of the the intelligent
systemsystem and company
and company approach. approach sales weights, and the variables affecting
FIBRES
16& TEXTILES in Eastern Europe 2016, Vol. 24, 6(120) FIBRES & TEXTILES in Eastern Europe 2017, Vol. 25, 1(121)
25