Research Article
Prototype of the Near-Infrared Spectroscopy Expert
System for Particleboard Identification

Anna Sandak ,1,2 Jakub Sandak ,1,2,3 Dominika Janiszewska,4 Salim Hiziroglu,5

Marta Petrillo,1 and Paolo Grossi1

1CNR-IVALSA Trees and Timber Institute, 38010 San Michele all’Adige, Italy
2InnoRenew CoE, 6310 Izola, Slovenia
3University of Primorska, Faculty of Mathematics, Natural Sciences and Information Technology, 6000 Koper, Slovenia
4Composite Wood Products Department, Wood Technology Institute, Poznan, Poland
5Natural Resource Ecology and Management, Oklahoma State University, Stillwater, Oklahoma, USA

Correspondence should be addressed to Jakub Sandak; jakub.sandak@innorenew.eu

Received 30 June 2018; Accepted 13 August 2018; Published 30 September 2018

Academic Editor: Jose S. Camara

Copyright © 2018 Anna Sandak et al. 'is is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

'e overall goal of this work was to develop a prototype expert system assisting quality control and traceability of particleboard
panels on the production floor. Four different types of particleboards manufactured at the laboratory scale and in industrial plants
were evaluated. 'e material differed in terms of panel type, composition, and adhesive system. NIR spectroscopy was employed
as a pioneer tool for the development of a two-level expert system suitable for classification and traceability of investigated
samples. A portable, commercially available NIR spectrometer was used for nondestructive measurements of particleboard panels.
Twenty-five batches of particleboards, each containing at least three independent replicas, was used for the original system
development and assessment of its performance. Four alternative chemometric methods (PLS-DA, kNN, SIMCA, and SVM) were
used for spectroscopic data classification. 'e models were developed for panel recognition at two levels differing in terms of their
generality. In the first stage, four among twenty-four tested combinations resulted in 100% correct classification. Discrimination
precision with PLS-DA and SVMC was high (>99%), even without any spectra preprocessing. SNV preprocessed spectra and
SVMC algorithm were used at the second stage for panel batch classification. Panels manufactured by two producers were 100%
correctly classified, industrial panels produced by different manufacturing plants were classified with 98.9% success, and the
experimental panels manufactured in the laboratory were classified with 63.7% success. Implementation of NIR spectroscopy for
wood-based product traceability and quality control may have a great impact due to the high versatility of the production and wide
range of particleboards utilization.

1. Introduction

Particleboard is a panel product manufactured from lig-
nocellulosic materials, combined with an adhesive system
and bonded together under heat and pressure. Particle-
boards are easy to process and flexible in application. 'ey
are mainly used for furniture production, but in combi-
nation with other materials, they might be used for parquet,
insulation materials, sheathing boards for timber framed
walls, packaging, or “do-it-yourself products.” 'e major
types of particles used to manufacture particleboard include
wood shavings, flakes, wafers, chips, and sawdust [1].

A new strategy of EU aims at increasing the share of
bioenergy in the EU’s total energy production. It is expected
that 20% of the energy produced in Europe after 2020 will be
from renewable resources, where 80% of the above will be
related to lignocellulosic feedstock. Simultaneously, a sig-
nificant increase in the production volumes for other wood-
based products (pulp, boards, and furniture) is expected.
'erefore, even if supply and demand of wood on the market
is balanced at present, it will most probably not be in
equilibrium within the coming years [2]. As a consequence,
several alternative resources, mainly agricultural and in-
dustrial residues, fast-growing shrubs and plantations, and

Hindawi
Journal of Spectroscopy
Volume 2018, Article ID 6025163, 11 pages
https://doi.org/10.1155/2018/6025163

mailto:jakub.sandak@innorenew.eu
http://orcid.org/0000-0002-2515-0991
http://orcid.org/0000-0001-9190-677X
https://creativecommons.org/licenses/by/4.0/
https://doi.org/10.1155/2018/6025163


postconsumer wood, have received considerable attention in
recent years [3]. Effective use of bagasse [4], oil palm waste
[5], bark [6] paper sludge, [3], kenaf stalks [7] wheat straw
[8], needle litter [9], vine pruning’s [10] waste tissue paper,
and corn peel [11] among others was previously reported.
Such materials are ecological, functional, and
environmental-friendly; therefore, they serve as sustainable
raw materials for manufacturing particleboards. 'e type of
raw resources used for panels manufacturing, its quality, size
of particles, moisture content as well as type and amount of
bonding system have significant effect on particleboard
properties [12]. 'e most important quality assessment
aspects of manufactured panels are emissions, which mainly
depends on the type and amount of resin [1]. 'e recent
trend to reduce the formaldehyde release from manufac-
tured wood products has led to the substitution of urea-
formaldehyde (UF) resin with several alternatives [12].
Liquefied wood was previously reported as an interesting
replacement for commonly used adhesives [13–15]. 'e
influence of liquefied wood on particleboard properties was
previously investigated by the authors [16, 17]. However,
alternative quality control tools with potential application
for online process control are still desired.

Portable spectroscopic equipment operating in the NIR
range is a highly interesting technology for the wood-based
sector [18]. Meder et al. [19] reported successful imple-
mentation of FT-NIR spectroscopy for at-line measurement
for quality control of melamine-urea-formaldehyde resin in
composite wood-panel production. Taylor and Via [20] used
visible and near-infrared spectroscopy to quantify phenol
formaldehyde resin content in oriented strandboard.
Campos et al. [21] used FT-NIR to evaluate composition of
agro-based particleboards. Janiszewska et al. [22] reported
the application of FT-NIR spectroscopy for clustering raw
materials used for liquefaction and their transformation
products. Even though several examples of NIR use for
product and process quality monitoring have been reported,
the majority of them rely on FT-NIR instruments that are
relatively costly (both time and investment wise) and with
limited applicability for inline measurement.

Numerous chemometric techniques for multivariate
classification and discrimination were reported as suitable
for processing of NIR spectra. 'ese are usually divided
into discriminant or class-modelling methods. In dis-
criminant analysis, an unknown sample is always assigned
and can be allocated only to one of the classes given in the
training set. 'e class-modelling approach is more flex-
ible. 'e sample can be accepted by more than one class
model and therefore be recognized as confused. 'e most
common classification methods are linear discriminant
analysis (LDA), partial least-squares discriminant analysis
(PLS-DA), k-nearest neighbours (kNN), SIMCA, and
SVM [23].

LDA is a probabilistic method, which assumes that each
sample belonging to a particular class follows a multivariate
Gaussian distribution. It requires the explicit calculation of
this probability for the formulation of the classification rule.
'e main disadvantage of this method is the fact that it
requires a significantly higher number of training samples

than the number of variables, the variables themselves
cannot be correlated, and all categories have the same
within-class scatter, which is hard to assure in many ex-
perimental cases [23].

PLS-DA is a multivariate inverse least-squares analysis
method used to classify samples. It decomposes the spectra
as linear combinations of principal components (PCs) that
express the majority of the information (variability) con-
tained in the global dataset. 'e predictor variables or latent
variables (LVs) are generated from the input variables to
maximize the variance between sample classes in the model
[24].

KNN is nonparametric and instance-based algorithm.
'is means that it does not make any assumption on the
underlying data distribution and does not use the training
data to do generalization. 'e kNN makes decision based on
the entire training data set. 'e object is classified by
a majority vote of neighbours. According to Adeniyi et al.
[25], the main advantages of this method are capability of
handling training data that are too large to fit in memory, use
of simple Euclidean distance to measure the similarities
between training and test data, providing a faster and more
accurate recommendation as a result of straightforward
application.

Soft independent modelling of class analogy (SIMCA) is
the most common supervised modelling method, repre-
senting class-modelling approach. It requires a training data
set of samples with a set of attributes and their class
membership. In SIMCA, a principal component analysis
(PCA) is performed on each class in the data set; thus,
a principal component model is used to represent each class.
In SIMCA, there is no restriction on the number of mea-
surement variables, and few samples per class are enough to
run the model.

Data classification is a common task in machine
learning. 'e support vector machine (SVM) is a very
flexible method that makes no assumption regarding data. It
is a nonlinear classification method that constructs a set of
hyperplanes in a high- or infinite-dimensional space. Good
separation is achieved by the hyperplane that has the largest
distance to the nearest training data point of any class [24]. It
works by obtaining the optimal boundary of two groups in
a vector space independent of the probabilistic arrangements
of vectors in training set. When the linear boundary in low-
dimension input space is not enough to separate two classes,
SVM can create a hyperplane that allows linear separation in
the higher dimension feature space [26]. SVM has been
successfully used for data mining, pattern recognition, and
artificial intelligence fields [18].

Practical implementation of discrimination methods in
the wood-based industry is a challenging task. Wood is
a heterogeneous and anisotropic biological material. 'e
correct implementation of NIR spectroscopy for bio-based
materials requires a complex approach [27]. A two-level
expert system is proposed here for identification of parti-
cleboard panels with a portable, commercially available NIR
spectrometer. 'e overall goal of this work was to develop
a prototype system that might assist quality control and
traceability of particleboard panels on the production floor.

2 Journal of Spectroscopy



2. Materials and Methods

Four different types of particleboards manufactured in the
laboratory and in industrial plants were evaluated. 'e
materials differed in terms of panel type (single-layer or
three-layer), composition (industrial particles, recycled
wood, or alternative lignocellulosic materials including fast-
growing species), and the adhesive system (urea-
formaldehyde (UF) resin, UF resin modified with lique-
fied wood (LW), or UF resin modified with starch). A
summary of investigated panels is presented in Table 1. In
total, 170 samples representing 25 variants were in-
vestigated. A colour image of representative samples surface
for each variant is presented in Figure 1.

2.1. Particleboards Manufactured with Recycled Wood (D).
Industrial wood particles obtained from a local wood-
processing sawmill, containing 25% recycled wood con-
tent, were used as a raw material for particleboard pro-
duction. 'e particles were sorted using an Allgaier
vibration screening machine with screens of mesh diameters
8, 2, 1, and 0.5 mm. Particle fractions ≤8 mm and ≥1 mm
were selected for particleboard production.

'e adhesive was prepared as a mixture of an industrial
urea-formaldehyde resin and liquefied wood (10–20% rel-
ative to the dry weight of the resin). 'e liquefied wood was
prepared from four types of wood-processing industry
wastes: mixed hardwood-softwood powder (LWP), pine
(LP), beech sawdust (LB), and bark (LB) [16].

'e industrial urea-formaldehyde glue resin character-
istics were as follows: gel time of 75 seconds, viscosity
336 mPa·s, total solids content 69.4%, and pH 7.3. Urea-
ammonium nitrate solution (46%) was used as a curing
agent, constituting 1% of the resin dry mass.

Single-layered particleboards of 12 mm thickness with 0,
10, 15, and 20% liquefied wood content in the adhesive resin
were produced in the laboratory. 'e nominal density of the
panels was 650 kg·m−3. All boards were conditioned after
pressing at 20°C and 65% relative humidity. Nine types of
single-layer panels were prepared with different adhesives
system, as listed in Table 1. Each panel type was produced in
two independent batches. Six replicates for each panel were
analysed; in total, 108 samples were investigated.

2.2. Particleboards Manufactured with Fast-Growing
Wood Species (S). Single- and three-layer particleboard
panels were manufactured from Eastern red cedar
(Juniperus virginiana L.) using 9% urea-formaldehyde, or
a combination of 15% modified corn starch and 2%
urea-formaldehyde adhesive. 'ree types of samples were
prepared: type A: single-layer board with 9% UF; type B:
three-layer board with 9% UF; and type C: three-layer
board with a combination of 15% starch and 2% UF. All
boards were produced in a laboratory press using a pressure
of 5 MPa, at a temperature of 165°C for 5 and 10 minutes in
the case of modified corn starch bonded samples [28]. Six
independent replicates for each panel were analysed with
an NIR spectrometer.

2.3. Particleboards Manufactured with Alternative to Wood
Lignocellulosic Plants (G). 'ree-layer particleboards with
the core made from different biomasses were produced,
including Black locust (Robinia pseudoacacia L.), miscant
(Miscanthus sinensis giganteus), willow (Salix viminalis), and
rapeseed (Brassica napus). 'e stalks of lignocellulosic
materials were reduced in size with a Pallmann’s chipper.
Fractions smaller than 10 mm and bigger than 1 mm were
used for the core of the panels. 'e chips were used for
manufacturing four types of three-layer boards of 16 mm
thickness, with the raw density of 680 kg·m−3 [29]. Six in-
dependent replicates for each panel were analysed with an
NIR spectrometer.

2.4. Industrially Manufactured Particleboards (P).
'ree-layer urea-formaldehyde resin-bonded particle-
boards, type P2, suitable for non-load-bearing purposes in
dry areas, manufactured by six diverse manufacturing plants
of a corporation were used as reference industrial samples.
In addition, panels with different thicknesses (38 mm,
28 mm, 18 mm, and 8 mm) prepared by a single producer
were investigated. 'ree replicates for each panel type were
analysed using NIR spectroscopy.

2.5. Spectroscopic Measurement and Data Mining. 'e
MicroNIR 1700 compact sensor produced by Viavi So-
lutions (Santa Monica, CA, USA) was used for spectro-
scopic measurements. Each spectrum was measured as an
average of 10 consecutive scans. 'e scanning fre-
quency was 50 Hz, corresponding to 20 ms of integration
time. 'e spectral range was from 950 to 1650 nm
(10526–6060 cm−1). 128 spectral points were defined for
each spectrum and corresponded to pixels of the CCD
detector of the instrument. Ten independent measure-
ments were done on each panel assuring measurements at
different locations over its surface. Each set of 10 spectra
was preprocessed with extended multiplicative scatter
correction (EMSC) and after that averaged to homogenize
the spectral fingerprint of heterogeneous surfaces of panel.
'e MicroNIR instrument has the proven potential for in-
field and inline applications due to its rigid construction
and integration of all optical, electronic, and mechanical
components.

PLS_Toolbox 8.0 (Eigenvector Research, Manson, WA,
USA) and LabView 13 (National Instruments, Austin TX,
USA) were used for data processing and mining. Four
discriminant analysis methods: PLS-DA, SVM, KNN, and
SIMCA, were used for spectra classification. Models were
calculated considering use of the raw spectra and spectra
preprocessed with normalization, standard normal variate
(SNV), EMSC, and 1st and 2nd derivatives (Savitzky–Golay
algorithm, 2nd polynomial order, and 15 smoothing points).
'e order of data was randomized and then divided into
calibration (66%) and independent validation (34%) data
sets. 'e samples used for model validation were all different
from those used for calibration.

Journal of Spectroscopy 3



3. Results and Discussion

3.1. General Concept of the Expert System Architecture.
Expert systems are innovative process tools that enhance the
user’s productivity in accomplishing a task or solving
a problem. 'ese vary in at least two important dimensions,
including knowledge and technological complexities [30].
Knowledge complexity is mainly determined by the degree
of depth and specialization of the internalized knowledge,
the scope of the decision, and the level of expertise required
to solve the problem. 'e technological complexity includes
diversity of hardware and software, the complexity of the
user’s environment, the scale of the software design effort,
the complexity of required database accesses, and special
user interfaces among others [30].

'e main goal of this work was to develop an expert system
capable of detecting and tracing different particleboards. 'e
general concept of such a system implemented in the furniture

factory is presented in Figure 2. 'e system relies on the
spectroscopic measurements acquired during arrival of the new
batch to the factory floor with the easy-to-handle and portable
instrument. 'e particular characteristics of the instrument
investigated within this project allow its direct implementation
in the real-life applications. It is recommended to average
several NIR spectra collected from the characterized sample in
order to minimize the heterogeneity effect of the complex
surfaces of particleboards. A spectrum acquired in that way is
considered as a fingerprint of the distinct batch and is recorded
in the database for further analysis. 'e usage of spectra is
twofold. Firstly, it is applied for the further improvement of the
chemometric models by providing new case data. Secondly, it is
used for future tracking and identification of elements during
production as well as in final products.

'e NIR spectrum of an unknown sample is confronted at
the first level of the expert system with broad classes repre-
senting different particleboard manufacturers or panel types.

Table 1: Manufactured panels investigated in this research.

Source
Sample
code

Manufacturing Panel type Raw material Adhesive system

Project LIDER

D1 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
0% LW + 100% UF1

D2 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
10% LWP +90% UF1

D3 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
15% LWP + 85% UF1

D4 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
20% LWP + 80% UF1

D5 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
20% LP + 80% UF1

D6 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
20% LB + 80% UF1

D7 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
20% LBK + 80% UF1

D8 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
0% LW + 100% UF2

D9 Laboratory Single-layer
Industrial particles containing 25%

of recycled wood
10% LWP + 90% UF2

OSU
S1 Laboratory Single-layer 100% red cedar particles 9% UF
S2 Laboratory 'ree-layer 100% red cedar particles 9% UF
S3 Laboratory 'ree-layer 100% red cedar particles 15% starch and 2% UF

ITD

G1 Laboratory 'ree-layer
Industrial pine particles (75%) + black

locust particles (25%)
Core 8% UF, face layers

12%

G2 Laboratory 'ree-layer
Industrial pine particles (75%)
+ Miscanthus particles (25%)

Core 8% UF, face layers
12%

G3 Laboratory 'ree-layer
Industrial pine particles (75%) + willow

particles (25%)
Core 8% UF, face layers

12%

G4 Laboratory 'ree-layer
Industrial pine particles (75%) + rapeseed

particles (25%)
Core 8% UF, face layers

12%

Chipboard panel
producer

P1 Industry 'ree-layer 100% industrial particles UF
P2 Industry 'ree-layer 100% industrial particles UF
P3 Industry 'ree-layer 100% industrial particles UF
P4 Industry 'ree-layer 100% industrial particles UF
P5 Industry 'ree-layer 100% industrial particles UF
P6a Industry 'ree-layer 100% industrial particles UF
P6b Industry 'ree-layer 100% industrial particles UF
P6c Industry 'ree-layer 100% industrial particles UF
P6d Industry 'ree-layer 100% industrial particles UF

4 Journal of Spectroscopy



In the case of successful identi�cation, the same spectrum is
an input for the second-level chemometric model, where
a	nity to the speci�c batches recorded previously is assessed.
Such a numerical tool might be highly useful for the quality
control of panels arriving to the production �oor, as well as
for authentication of their origin and composition. �e de-
monstrative implementation of the proposed expert system
was performed within this research project. �e statistical
evaluation of the discriminations success rate, performed at
both system levels, is presented in the following section.

3.2. Prototype Expert System for Particleboard Identi�cation.
�e concept of the abovementioned expert system imple-
mented for demonstration of its feasibility is presented in

Figure 3. It consists of �ve chemometric models imple-
mented at two discrimination levels. Model #1 is used to
screen the unknown spectrum and determine the most
probable class corresponding to the particleboard type.
Four classes were therefore de�ned at the �rst level of the
expert system corresponding to “sample sources” as
summarized in the �rst column of Table 1. �e second and
more speci�c discrimination is executed afterward by
implementing one of the second-level models (#2 to #5).
�e number of classes in each of these models corresponds
to the number of previously de�ned batches. It has to be
mentioned that it is critical to continuously feed the expert
system with the most recent spectra of new batches in order
to assure constant improvement of the discrimination

D1 D2 D3 D4 D5

D6 D7 D8 D9 G1

S1 S2 G2 G3 G4

S3 P1 P2 P3 P4

P5 P6a P6b P6c P6d

Figure 1: Appearance of particleboard surfaces representing 25 investigated categories.

Journal of Spectroscopy 5



system and proper identi�cation of all historically used
particleboard batches.

An expert system approach to quality control at the
industrial scale was proposed by Paladini [31]. He de-
scribed three-step quality control approach that included
precontrol activities, inspection, and a decision stage.
Liukkonen et al. [32] described the intelligent optimiza-
tion and modelling system for electronics production.
�ey proposed three modules consisting of appropriate
mathematical tools speci�cally tailored to each task:
preprocessing, variable selection, and optimization. A
data-driven approach was proposed to achieve proactive
quality improvement of the production process. Hob-
ballah et al. [33] proposed a casual map to show what
variables should be included in the design optimization
and how the components interact causally. �is approach

was successfully used for the preliminary design of an
insulating composite mat based on wood �bers.

3.3. Discriminant Analysis. Four alternative classi�cation
methods: KNN, PLS-DA, SVM, and SIMCA, were tested
for the optimal NIR spectral data classi�cation. All these
methods are supervised techniques, as they use prede�ned
information about the class membership for all samples
selected for model calibration. In that way, the model can
be tuned to classify new unknown samples in one of the
known classes on the basis of its individual pattern [26].
Discriminating techniques usually build models based on
all the categories concerned for the discrimination.
�erefore, samples can be classi�ed into one of the pre-
de�ned categories, even if actually they do not belong to

Production of
particleboards

(raw resources, 
adhesives, and 
technologies)

Arrival at the warehouse
(registering batches 
and quality control)

Storage of panels

Processing of panels

Assembly and quality control

Use phase
(complaints and guarantee 

management)

End of life
(recycling, reuse, 

and circular economy)

Database
ERP/MRP system

Batch #001

Logistics

...

Producer #A Producer #B Producer #n

NIR fingerprint for batch #A001

#A001
#B401

#C055

#A002 #D016 #M099

Product #1 Product #2 Product #n

...

Client X

Figure 2: Concept for the traceability of particleboards in the furniture industry.

6 Journal of Spectroscopy



any of these [26]. It is possible, however, to de�ne a certain
threshold for a	nity to any of the classes. In that case,
samples not passing the threshold may be classi�ed as
unde�ned.

Particleboard panel discrimination at the producer level
(model #1) di�erentiated classes in terms of manufacturing
conditions (laboratory or industrial scale), di�erent re-
sources used for panels manufacturing (industrial particles,
addition of recycled wood, fast-growing wood species, ag-
ricultural wastes, etc.), or di�erent adhesives used for panel
production. �e discrimination success scores between
di�erent panel providers are summarized in Table 2, where
selected spectra preprocessing and diverse discrimination
algorithms are presented. Four among twenty-four tested
combinations resulted in 100% correct classi�cation. �e
optimal combinations were PLS-DA and SIMCA with 2nd

derivative spectra, and SVM with both SNV and EMSC
preprocessing. Furthermore, discrimination precision with
PLS-DA and SVM was high (>99%), even without any
spectra preprocessing.

�e second level of the expert system was designed to
identify speci�c batches. Twenty-�ve lots of particleboards
were modelled by four independent chemometric models
(#2 to #5), developed for each panel provider separately.
�e SVM algorithm classifying SNV preprocessed NIR
spectra was implemented. �e summary of the model’s
performance is presented in Figure 4, where confusion
tables indicate the ability of each model to properly classify
(or misclassify) experimented samples from the validation
set. All samples of panels manufactured by producer S and
G were correctly classi�ed. Industrial panels produced by
di�erent manufacturing plants of the producer P were
classi�ed with 98.9% success and the panels manufactured
by producer D with 63.7% success. It is important to
mention that, in the case of the producer D, the same
substrate (industrial particles containing 25% of recycled

wood) was used for manufacturing all investigated panels.
�e only di�erence was the adhesive system, with 0, 10, 15,
and 20% share of lique�ed wood addition in the UF resin.

3.4. Implementation of the Prototype Expert System. �e
implementation of the expert system in real conditions is
a challenging task. Qian et al. [34] proposed a four-step
process that consists of the following:

(i) Knowledge representation (including all variables
characterizing production process)

(ii) Database development (including all decision rules,
functions, and metadata)

(iii) Machine interface design (in�uencing the quality of
the expert system for fault diagnosis and the real-
time response of the system)

(iv) Knowledge maintenance (including update, veri�-
cation, and correction of the errors in knowledge
base)

�e experiences gained when developing chemometric
models discriminating diverse particleboards stimulated
extension of this research to a prototype expert system ready
for testing in real-world applications. �e hardware used for
the prototype (portable NIR spectrometer and computer)
was similar to the laboratory tests. �e di�erences were in
the software, where the original code was customised. It
included integration of tools for the control of the sensor
setup, data acquisition, spectra processing, and discriminant
analysis. �e required chemometric models (#1 to #5) were
computed o©ine with PLS_toolbox and were exported as
Matlab scripts to LabView. �e algorithm of the prototype
expert system for the batch identi�cation is presented in
Figure 5. �e software allows acquisition of the new NIR

Unknown particleboard sample

(at least 10) NIR spectra measurements

Spectra homogenization (EMSC + average)

Model #1
(panel supplier)

Model #2
(project Lider-D)

Model #3
(OSU-S)

Model #4
(ITD-G)

Model #5
(industry-P)

D1 D2 D3 D4 D5 D6 D7 D8 D9 S1 S2 S3 G1 G2 G3 G4 P1 P2 P3 P4 P5 P6-a P6-b P6-c P6-d

Figure 3: Prototype expert system’s layout for the experimental set of particleboard samples.

Journal of Spectroscopy 7



spectra for feeding the database as well as assessment of the
unknown spectrum in one of the prede�ned classes.

4. Conclusions

�e traceability of particleboards is an important practical
issue a�ecting economic performance of furniture producers
and related industries. It allows control of material quality
and ensures high standards of �nal products. NIR spec-
troscopy was employed here as a pioneer tool for devel-
opment of the expert system suitable for assisting such
a traceability. Twenty-�ve batches of particleboards, each
containing at least three independent replicas, were used for
the original system development and for its performance
assessment. �e hardware and software developed was

working properly, and it was implemented as a ready-to-use
prototype system.

A two-stage expert system seems to be optimal for
implementation of the particleboard discrimination as it
allows a rough (in the �rst stage) and detailed (in the second
stage) classi�cation of particleboard samples to the level of
a single batch. In two cases (two producers), 100% correct
classi�cation was achieved at the second stage.

It has to be mentioned that a drawback of the system
presented is the necessity to condition samples before
measurement. It includes both refreshing of the surface
by gentle sanding and conditioning of samples at well-
de�ned climatic conditions. It is due to the fact that NIR
spectroscopy is extremely sensitive to the water mole-
cules presented in the measured samples. Nevertheless,

Predicted as D1
Predicted as D3

Predicted as D5
Predicted as D7

Predicted as D9

0
10
20
30
40
50
60
70
80
90

100

D1

N
um

be
r o

f o
bs

er
va

tio
n 

(%
)

D2 D3 D4 D5 D6 D7 D8 D9

(a)

Predicted as S-A
Predicted as S-B

Predicted as S-C

0
10
20
30
40
50
60
70
80
90

100

S-A S-B S-C

(b)

Predicted as GM
Predicted as GR

Predicted as GS
Predicted as GW

0
10
20
30
40
50
60
70
80
90

100

GM GR GS GW

(c)

Predicted as P-1
Predicted as P-3

Predicted as P-5
Predicted as P-6-b

Predicted as P-6-d

0

P-
1

P-
2

P-
3

P-
4

P-
5

P-
6-

a
P-

6-
b

P-
6-

c
P-

6-
d

10
20
30
40
50
60
70
80
90

100

(d)

Figure 4: Confusion table of the particleboard batch prediction with the prototype NIR expert system: project LIDER-D (a), OSU-S (b),
ITD-G (c), and industry-P (d).

Table 2: �e results of the discrimination success rate at the �rst level of expert system.

Discrimination algorithm
Spectra preprocessing

No preprocessing Normalization SNV EMSC 1st derivative 2nd derivative
PLS-DA 99.1 99.1 98.2 98.2 98.2 100
SVM 99.1 63.2 100 100 93.9 63.2
KNN [3] 93.0 96.5 97.4 96.5 97.4 98.2
SIMCA 94.7 99.1 99.1 98.2 99.1 100

8 Journal of Spectroscopy



implementation of NIR spectroscopy for product
traceability and quality control may impact the industry
due to the high versatility of the production and wide
range of particleboard use.

Data Availability

'e data used to support the findings of this study are
available from the corresponding author upon request.

Conflicts of Interest

'e authors declare that there are no conflicts of interest
regarding the publication of this paper.

Acknowledgments

'e authors are grateful to Dorota Fuczek, Grzegorz
Kowaluk, and Ekkehard Brommer for providing part of
experimental samples. Special acknowledgments are due to
COST Action FP1306 (LIGNOVAL) for funding an STSM
that contributed to this work. Part of the work was

conducted during the BIO4ever (RBSI14Y7Y4) project
funded within SIR call by the Italian Ministry of Education,
University and Research (MIUR). 'e authors gratefully
acknowledge the European Commission for funding the
InnoRenew CoE project (Grant Agreement #739574) under
the Horizon2020 Widespread-Teaming program. 'e re-
search was carried out within the statutory project “Ap-
plication of liquefied wood in wood-based materials
technology, Part 3. 'e use of liquefied lignocellulosic wastes
as a binder for particleboards” (ST-2-BMD/2015/K) and
project LIDER/14/0174/L-7/15/NCBR/2016 “New bio-
polymer adhesives modified with silanes and ionic liquids
for application in wood-based materials technology” funded
by the NCBR under the LIDER VII programme.

References

[1] Compilation of Air Pollutant Emissions Factors, Chapter 10:
Wood Products Industry Wood Products Industry, 2002,
https://www3.epa.gov/ttn/chief/ap42/ch10/index.html, 2018.

[2] 'e Bioenergy System Planners Handbook—BISY PLAN,
http://bisyplan.bioenarea.eu/, 2018.

Start

Measure NIR spectra
of unknown sample

Compute model #2

Compute model #3

Compute model #4

Compute model #5

Compute model #1

Is producer = D

Is producer = S

Is producer = G

Is producer = P

Define batch #

Define batch #

Define batch #

Define batch #

The sample is
“Undefined”

Visualize the result

End

No

No

No

No

Figure 5: 'e algorithm of the prototype expert system.

Journal of Spectroscopy 9

https://www3.epa.gov/ttn/chief/ap42/ch10/index.html
http://bisyplan.bioenarea.eu/


[3] A. Taramian, K. Dooshoseini, S. A. Mirshokraii, and
M. Faezipour, “Particleboard manufacturing: an innovative
way to recycle paper sludge,” Waste Management, vol. 27,
no. 12, pp. 1739–1746, 2007.

[4] J. Fiorelli, R. G. Galo, S. L. Castro Jr., U. L. Belini,
P. R. O. Lasso, and H. Savastano Jr., “Multilayer particleboard
produced with agroindustrial waste and amazonian vegetable
fibres,” Waste and Biomass Valorization, vol. 9, no. 7,
pp. 1151–1161, 2017.

[5] L. S. Hua, Z. Ashaari, L. W. Chen et al., “Properties of par-
ticleboard with oil palm trunk as core layer in comparison to
three-layer rubber wood particleboard,” Journal of Oil Palm
Research, vol. 27, no. 1, pp. 67–74, 2015.

[6] M. C. N. Yemele, P. Blanchet, A. Cloutier, and A. Koubaa,
“Effects of bark content and particle geometry on the physical
and mechanical properties of particleboard made from black
spruce and trembling aspen bark,” Forest Product Journal,
vol. 58, no. 11, pp. 48–56, 2008.

[7] H. Kalaycioglu and G. Nemli, “Producing composite parti-
cleboard from kenaf (Hibiscus cannabinus L.) stalks,” In-
dustrial Crops and Products, vol. 24, no. 2, pp. 177–180, 2006.

[8] X. Mo, E. Cheng, D. Wang, and X. S. Sun, “Physical properties
of medium-density wheat straw particleboard using different
adhesives,” Industrial Crops and Products, vol. 18, no. 1,
pp. 47–53, 2003.

[9] G. Nemli, S. Yildiz, and E. D. Gezer, “'e potential for using
the needle litter of Scotch pine (Pinus sylvestris L.) as a raw
material for particleboard manufacturing,” Bioresource
Technology, vol. 99, no. 14, pp. 6054–6058, 2008.

[10] G. A. Ntalos and A. H. Grigoriou, “Characterization and
utilisation of vine prunings as a wood substitute for parti-
cleboard production,” Industrial Crops and Products, vol. 16,
no. 1, pp. 59–68, 2002.

[11] P. Lertsutthiwong, S. Khunthon, K. Siralertmukul,
K. Noomun, and S. Chandrkrachang, “New insulating par-
ticleboards prepared from mixture of solid wastes from tissues
paper manufacturing and corn peel,” Bioresource Technology,
vol. 99, no. 11, pp. 4841–4845, 2008.

[12] N. Čuk, M. Kunaver, and S. Medved, “Properties of particle-
boards made by using an adhesive with added liquefied wood,”
Materials and technology, vol. 45, no. 3, pp. 241–245, 2011.

[13] B. Esteves, J. Martins, J. Martins, L. Cruz-Lopes, J. Vicente,
and I. Domingos, “Liquefied wood as a partial substitute of
melamine-urea-formaldehyde and urea-formaldehyde
resins,” Maderas. Ciencia y tecnoloǵıa, vol. 17, no. 2,
pp. 277–284, 2015.

[14] A. Ugovsek, M. Kariz, and M. Sernek, “Bonding of wood with
adhesive mixtures made of liquefied wood combined with
tannin or phenolic resin,” in Future with Forests, R. Ristić,
M. Medarević, and Z. Popović, Eds., First Serbian Forestry
congress, Beograd, Serbia, 2010.

[15] M. Kunaver, S. Medved, N. Čuk, E. Jasiukaitytė, I. Poljanšek,
and T. Strnad, “Application of liquefied wood as a new particle
board adhesive system,” Bioresource Technology, vol. 101,
no. 4, pp. 1361–1368, 2010.

[16] D. Janiszewska, I. Frąckowiak, and K. Mytko, “Exploitation of
liquefied wood waste for binding recycled wood particle-
boards,” Holzforschung, vol. 70, no. 12, pp. 1135–1138, 2016.

[17] D. Janiszewska, I. Frąckowiak, and N. Bielejewska, “Appli-
cation of selected agents for wood liquefaction and some
properties of particleboards produced with the use of liquefied
wood,” Drewno, vol. 59, no. 197, pp. 223–230, 2016.

[18] J. Sandak, A. Sandak, and R Meder, “Assessing trees, wood
and derived products with NIR spectroscopy: hints and tips,”

Journal of Near Infrared Spectroscopy, vol. 24, no. 6,
pp. 485–505, 2016.

[19] R. Meder, W. Stahl, P. Warburton et al., “At-line validation of
a process analytical technology approach for quality control of
melamine-urea-formaldehyde resin in composite wood-panel
production using near infrared spectroscopy,” Analytical and
Bioanalytical Chemistry, vol. 409, no. 3, pp. 763–771, 2017.

[20] A. Taylor and B. K. Via, “Potential of visible and near infrared
spectroscopy to quantify phenol formaldehyde resin content
in oriented strandboard,” European Journal of Wood and
Wood Products, vol. 67, no. 1, pp. 3–5, 2009.

[21] A. C. M. Campos, P. R. G. Hein, R. F. Mendes, L. M. Mendes,
and G. Chaix, “Near infrared spectroscopy to evaluate
composition of agro-based particleboards,” Bioresources,
vol. 4, no. 3, pp. 1058–1069, 2009.

[22] D. Janiszewska, A. Sandak, J. Sandak, I. Frąckowiak, and
K. Mytko, “Influence of the raw material properties on the
liquefied wood chemical composition,” nnals of WULS,
Forestry and Wood Technology, vol. 94, pp. 298–303, 2016.

[23] M. Bevilacqua, R. Nescatelli, R. Bucci, A. D. Magr̀ı,
A. L. Magr̀ı, and F. Marini, “Chemometric classification
techniques as a tool for solving problems in analytical
chemistry,” Journal of AOAC International, vol. 97, no. 1,
pp. 19–28, 2014.

[24] Y. Tian, Z. Wang, X. Han, H. Hou, and R. Zheng, “Com-
parative investigation of partial least squares discriminant
analysis and support vector machines for geological cuttings
identification using laser-induced breakdown spectroscopy,”
Spectrochimica Acta Part B: Atomic Spectroscopy, vol. 102,
pp. 52–57, 2014.

[25] D. A. Adeniyi, Z. Wei, and Y. Yongquan, “Automated web
usage data mining and recommendation system using
K-nearest neighbor (KNN) classification method,” Applied
Computing and Informatics, vol. 12, no. 1, pp. 90–108, 2016.

[26] L. A. Berrueta, R. M. Alonso-Salces, and K. Héberger, “Su-
pervised pattern recognition in food analysis,” Journal of
Chromatography A, vol. 1158, no. 1-2, pp. 196–214, 2007.

[27] J. Liu, T. J. Pan, and Z. Y. Zhang, “Incremental support vector
machine combined with ultraviolet-visible spectroscopy for
rapid discriminant analysis of red wine,” Journal of Spec-
troscopy, vol. 2018, Article ID 4230681, 5 pages, 2018.

[28] J. Sandak, A. Sandak, S. Hiziroglu, and S. Bosak, “Surface
characterization of particleboard panels manufactured from
eastern redcedar using multi-sensor approach,” in Proceedings
of International Panel Products Symposium, pp. 27–37,
Llandudno, North Wales, UK, October 2015.

[29] I. Frąckowiak, D. Fuczek, and G. Kowaluk, “Impact of dif-
ferent lignocellulosic materials used in core of particleboard
on modulus of elasticity and bending strength,” Drewno,
vol. 51, no. 180, pp. 5–14, 2008.

[30] J. H. Bradley and R. D. Hauser, “A framework for expert
system implementation,” Expert Systems with Applications,
vol. 8, no. 1, pp. 157–167, 1995.

[31] E. P. Paladini, “An expert system approach to quality control,”
Expert Systems with Applications, vol. 18, no. 2, pp. 133–151,
2000.

[32] M. Liukkonen, E. Havia, H. Leinonen, and Y. Hiltunen,
“Expert system for analysis of quality in production of
electronics,” Expert Systems with Applications, vol. 38, no. 7,
pp. 8724–8729, 2011.

[33] M. H. Hobballah, A. Ndiaye, F. Michaud, and M. Irle,
“Formulating preliminary design optimization problems us-
ing expert knowledge: application to wood-based insulating

10 Journal of Spectroscopy



materials,” Expert Systems with Applications, vol. 92, pp. 95–
105, 2018.

[34] Y. Qian, L. Xu, X. Li, L. Lin, and A. Kraslawski, “LUBRES: an
expert system development and implementation for real-time
fault diagnosis of a lubricating oil refining process,” Expert
Systems with Applications, vol. 35, no. 3, pp. 1252–1266, 2008.

Journal of Spectroscopy 11



Tribology
Advances in

Hindawi
www.hindawi.com Volume 2018

Hindawi
www.hindawi.com Volume 2018

 International Journal ofInternational Journal of

Photoenergy

Hindawi
www.hindawi.com Volume 2018

Journal of

Chemistry

Hindawi
www.hindawi.com Volume 2018

Advances in
Physical Chemistry

Hindawi
www.hindawi.com

 Analytical Methods  
in Chemistry

Journal of

Volume 2018

Bioinorganic Chemistry 
and Applications
Hindawi
www.hindawi.com Volume 2018

Spectroscopy
International Journal of

Hindawi
www.hindawi.com Volume 2018

Hindawi Publishing Corporation 
http://www.hindawi.com Volume 2013
Hindawi
www.hindawi.com

The Scientific 
World Journal

Volume 2018

Medicinal Chemistry
International Journal of

Hindawi
www.hindawi.com Volume 2018

Nanotechnology
Hindawi
www.hindawi.com Volume 2018

Journal of

Applied Chemistry
Journal of

Hindawi
www.hindawi.com Volume 2018

Hindawi
www.hindawi.com Volume 2018

Biochemistry 
Research International

Hindawi
www.hindawi.com Volume 2018

Enzyme 
Research

Hindawi
www.hindawi.com Volume 2018

Journal of

SpectroscopyAnalytical Chemistry
International Journal of

Hindawi
www.hindawi.com Volume 2018

Materials
Journal of

Hindawi
www.hindawi.com Volume 2018

Hindawi
www.hindawi.com Volume 2018

BioMed 
Research International Electrochemistry

International Journal of

Hindawi
www.hindawi.com Volume 2018

N
a

n
o

m
a

te
ri

a
ls

Hindawi
www.hindawi.com Volume 2018

Journal of

Nanomaterials

Submit your manuscripts at
www.hindawi.com

https://www.hindawi.com/journals/at/
https://www.hindawi.com/journals/ijp/
https://www.hindawi.com/journals/jchem/
https://www.hindawi.com/journals/apc/
https://www.hindawi.com/journals/jamc/
https://www.hindawi.com/journals/bca/
https://www.hindawi.com/journals/ijs/
https://www.hindawi.com/journals/tswj/
https://www.hindawi.com/journals/ijmc/
https://www.hindawi.com/journals/jnt/
https://www.hindawi.com/journals/jac/
https://www.hindawi.com/journals/bri/
https://www.hindawi.com/journals/er/
https://www.hindawi.com/journals/jspec/
https://www.hindawi.com/journals/ijac/
https://www.hindawi.com/journals/jma/
https://www.hindawi.com/journals/bmri/
https://www.hindawi.com/journals/ijelc/
https://www.hindawi.com/journals/jnm/
https://www.hindawi.com/
https://www.hindawi.com/