Contents lists available at ScienceDirect
Journal of Advanced Research
Original article
A regression-tree multilayer-perceptron hybrid strategy for the
prediction of ore crushing-plate lifetimes
Mario Juez-Gila, Ivan Nikolaevich Erdakovb, Andres Bustilloa, Danil Yurievich Pimenovc,
a Department of Civil Engineering, Universidad de Burgos, Avda Cantabria s/n, Burgos 09006, Spain
b Foundry Department, South Ural State University, Lenin Prosp. 76, Chelyabinsk 454080, Russia
c Department of Automated Mechanical Engineering, South Ural State University, Lenin Prosp. 76, Chelyabinsk 454080, Russia
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
 Dataset of plates lifetime were
obtained by 3 casting methods and
chemical composition.
 A two-steps model for prediction of
the full lifetime of plates of Hadfield
steel was proposed.
 The prediction model combines
regression trees with multilayer
perceptron (MLP)
 MLP provides accurate wears models
considering the chemical
 Regression trees provide visual
information about dataset structure
to build MLP.
a r t i c l e
i n f o
a b s t r a c t
Article history:
Highly tensile manganese steel is in great demand owing to its high tensile strength under shock loads.
Received 15 December 2018
Revised 21 March 2019
Accepted 21 March 2019
Available online 23 March 2019
All workpieces are produced through casting, because it is highly difficult to machine. The probabilistic
aspects of its casting, its variable composition, and the different casting techniques must all be considered
for the optimisation of its mechanical properties. A hybrid strategy is therefore proposed which combines
decision trees and artificial neural networks (ANNs) for accurate and reliable prediction models for ore
Hadfield steel
Resource savings
Lifetime prediction
Regression trees
crushing plate lifetimes. The strategic blend of these two high-accuracy prediction models is used to gen-
erate simple decision trees which can reveal the main dataset features, thereby facilitating decision-
making. Following a complexity analysis of a dataset with 450 different plates, the best model consisted
of 9 different multilayer perceptrons, the inputs of which were only the Fe and Mn plate compositions.
The model recorded a low root mean square error (RMSE) of only 0.0614 h for the lifetime of the plate:
Multi-layer perceptrons
a very accurate result considering their varied lifetimes of between 746 and 6902 h in the dataset. Finally,
Artificial intelligence
the use of these models under real industrial conditions is presented in a heat map, namely a 2D repre-
sentation of the main manufacturing process inputs with a colour scale which shows the predicted out-
put, i.e. the expected lifetime of the manufactured plates. Thus, the hybrid strategy extracts core training
Peer review under responsibility of Cairo University.
Corresponding author.
E-mail address: danil_u@rambler.ru (D.Y. Pimenov).
2090-1232/ 2019 The Authors. Published by Elsevier B.V. on behalf of Cairo University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
M. Juez-Gil et al./Journal of Advanced Research 18 (2019) 173–184
dataset information in high-accuracy prediction models. This novel strategy merges the different capabil-
ities of two families of machine-learning algorithms. It provides a high-accuracy industrial tool for the
prediction of the full lifetime of highly tensile manganese steel plates. The results yielded a precision pre-
diction of (RMSE of 0.061 h) for the full lifetime of (light, medium, and heavy) crusher plates manufac-
tured with the three (experimental, classic, and highly efficient (new)) casting methods.
 2019 The Authors. Published by Elsevier B.V. on behalf of Cairo University. This is an open access article
mum parameters, the new technology requires neither heavy
heads nor labour-intensive operations with the casting form both
Highly tensile manganese steel, also known as Hadfield steel,
before and after pouring to achieve the optimum angle; thereby
named after its first manufacturer, consisting of 11.5–15.0% of
decreasesing the cost of producing plates and leading to consider-
Mn and 0.9–1.4% of C demonstrates high tensile strength under
able savings on metal in the gating system and machine heads (15–
shock loads, such as in tank track operation, tractors and other
machines, bucket tooth
bars for
limestone, ore
As we approach the fourth technological revolution in the set-
crusher jaws, and railroad track switches on wheel sets. The afore-
ting of global competition, the analysis of all existing data from
mentioned properties are due to the interaction of steel with a
the casting process becomes increasingly relevant in terms of iden-
softer material and the absence of scuffing on the impact surface
tifying the best strategies which will optimise the mechanical
of the steel workpiece, thus causing fatigue-induced rather than
characteristics, particularly the wear resistance of components
abrasive wear. As a consequence of the difficulties associated with
which are cast using this steel type, thus creating a competitive
cutting this alloy, highly tensile manganese-steel workpieces are
advantage. Previously unknown and hidden trends can be useful,
typically produced via casting.
and comprehensible patterns found at the intersection of data-
Extensive research on improvements in this type of steel
bases, statistics, and machine-learning techniques. The size of the
reflects the active industrial interest in its mechanical properties.
database (big data or data of a specific experiment) is not essential;
Siafakas et al. [1] conducted a quantitative analysis of the amount,
the importance lies in the identification of hidden patterns, which
size, and number of particles which precipitate in situ in titanium-
would be impossible to establish with direct visual analysis or by
and aluminium-treated Hadfield steel during casting. In certain
calculating simple statistical features.
research works, heat treatment has been suggested as a means of
Casting is an inherently probabilistic process; the quality of a
increasing the micro-hardness of the cast Hadfield steel matrix
cast is primarily attributed to the chemical composition of the
[2–4]. Moreover, in several studies [5–7], the factors which can
alloy and the nature of its solidification. The objective of finding
affect the increased wear resistance of high-manganese steel have
hidden patterns in the array of technological data from the produc-
been examined.
tion and operation of steel plates used at crushing stations appears
Wear resistance appears to be the focus of most research efforts
relevant to the investigation of the reasons which cause their wear.
owing to the fact that it can extend the workpiece lifetime. There
Therefore, the objective of this study is to extend the total lifetime
are works dedicated to the study of wear resistance in high-
of (light, medium, and heavy) Hadfield steel plates for ore process-
speed pounding (HSP) of Hadfield steel to produce a thick
ing equipment by revealing new trends using machine-learning
nanocrystalline surface layer with gradient nanostructure [8].
techniques to model their wear limits.
Abbasi et al. [9] studied the abrasive wear behaviour of Al-
The solutions of complex industrial manufacturing processes, as
alloyed Hadfield steel under both high- and low-stress wear condi-
presented in this study, typically follow two separate strategies. In
tions in comparison with that of non-Al alloyed Hadfield steel.
the first one, the use of analytical models is proposed based on
Kolokoltsev et al. [10] studied the resistance of Hadfield steel
experimental data; in certain cases, this strategy is supported by
cooled at different rates. El-Fawkhry et al. and Kalandyk et al.
physical models or simulations of the manufacturing process and
[11,12] both discussed the results of austenitic matrix modification
is fine-tuned with the experimental data acquired under labora-
in high-manganese steel castings. Smith et al. [13] studied the
tory conditions. This approach has already been discussed in the
materials produced through the addition of minor amounts of
introduction for the prediction of the lifetime of ore crushing
other carbide-forming and solid-solution strengthening elements
plates. In the second approach, machine-learning techniques are
and through the heat treatment of the as-cast components under
employed to build prediction models from massive datasets; this
pressure. Te˛cza and Głownia and Głownia et al. [14,15] studied
approach could become a suitable tool for decision making.
the composite structure of high-manganese steel using vanadium
Each approach has its advantages and disadvantages. The ana-
carbides following melting and solidification. Najafabadi et al.
lytical models are typically based on homogeneous and simplified
[16] studied the wear resistance of cast Hadfield steel after adding
manufacturing processes, first, because they use data for fine-
Ti elements. Zhong et al. [17] studied the effect of the composite
tuning collected under restricted laboratory conditions (to reduce
structure of (Fe, Cr)7C3-Fe on its wear resistance and concluded
experimental costs); second, because they are meant to consider
that it was 1.34 times higher than that of the Hadfield steel. Finally,
only variables of the same nature in the manufacturing process,
Zhang et al. [18] examined a composite coating of WC/Hadfield
e.g. cutting conditions and chemical composition. However, they
steel produced via centrifugal casting to improve its impact wear
rarely mix variables of distinctly different natures, because the
analytical and physics-based models are not designed for such
However, all aforementioned methods complicate the technol-
tasks. The most common machine-learning approaches, such as
ogy of manufacturing workpieces using Hadfield steel and cause
artificial neural networks, belong to the black-box category of
it to be more expensive. Moreover, insufficient attention has been
these techniques, i.e. they provide no equation which shows the
paid to the issue of resource conservation, with the exception of
relationship between inputs and outputs. The only manner in
the studies by Erdakov et al. [19–22], who proposed a new highly
which the information contained in those models can be extracted
efficient gating and feeding system and defined its optimum
is to query the predicted output for a certain combination of inputs
parameters for casting using green sand moulds. With the opti-
and the prediction model will provide an estimated value. Hence, if
M. Juez-Gil et al./Journal of Advanced Research 18 (2019) 173–184
useful information would be extracted from these models, they
Research material and methods
would require either a 2D or a 3D representation of their predic-
tions [23–25]. This approach has been successfully validated for
Before developing a model for the prediction of the lifetime of
several industrial tasks, for example, in predicting surface rough-
crushing plates and prior to conducting an experiment, it is neces-
ness [26–28], surface quality [29], and cutting-tool wear [30,31],
sary to determine the properties of the materials that are used, the
among others. Furthermore, the datasets required to train these
parameters of the cast products, as well as the casting and investi-
models should be as big and diverse as possible. However, indus-
gation methods.
trial data are limited to real-life scenarios, given the reluctance of
the industry to finance tests which go beyond the specification of
manufacturing conditions. Nevertheless, such tests are essential
in the training process of machine-learning techniques. Moreover,
Plate manufacturing and casting methods
part of the information in the datasets, rather than relating to the
manufacturing problem, is related to the experimental design
method itself (e.g. if in a certain cutting process we test a range
of cutting tools, each having an additional tooth and an extra
5 mm diameter in addition to those of the preceding one, as per
the specifications of the manufacturer, then the machine-learning
model will conclude that the number of teeth and the diameter
of the tool are two completely correlated inputs, playing the same
role in the cutting process).
Although the most common machine-learning techniques
belong to the black-box category, there are certain machine-
learning techniques, such as decision trees, that provide visual
information on the process. However, these techniques are often
simpler than artificial neural networks (ANNs) and might not per-
form equally well in very complex processes, although they are free
from the complexity and tediousness of fine-tuning the ANN model
parameters.In thisstudy, we proposea hybridstrategyto overcome
this limitation, which combines decision trees for extraction of the
main information included in the training dataset with ANNs for
high-accuracy prediction models. This strategy combines the great-
est advantagesof both machine-learning techniques: to understand
the main features of the dataset, it generates rapid, visual, and sim-
ple decision trees, thereby facilitating decision-making on inputs
for simple, yet accurate, ANN models.
pre-analysis wasperformed usingreducederror pruning(REP)trees,
only eight chemical components as inputs for the prediction model.
Then, the 9 independent prediction models (one for each subset)
for 13 different multilayer perceptron (MLP) structures (the most
promising combinations of chemical components) were trained
and the most accurate models were identified. The test of only some
of the possible combinations of chemical composition of the ore
plates in the MLPs is an industrial requirement (to reduce the mod-
elling effort). Meanwhile, the efficient selection of the features used
in the training stage of the MLPs is an interesting challenge owing
to the high number of possible combinations. Then, the complexity
of the MLP structure was considered to select the best prediction
model from an industrial perspective. Finally, the identification of a
high-accuracypredictionmodel may beinsufficientforitssuccessful
implementation under real industrial conclusions. Therefore, the
industrial use, namely a 2D representation of the main inputs of the
manufacturing process with a colour scale showing the predicted
output, i.e. the wear limit of the manufactured plates.
This strategy is able to deal with data of different natures, the
chemical composition of the plates, and the manufacturing process
of the plates in our case study. Moreover, the strategy produces
models which are optimised in terms of accuracy, with a reduced
In this research, the following materials and research methods
were used. The chemical composition (%) of Hadfield steel is listed
in Table 1. Hadfield steel contains 84.3–87.3% iron (Fe), 11.5–15.0%
magnesium (Mn), 0.9–1.4% carbon (C), 0.3–1.0% silicon (Si), and 0–
3% impurities. The physical and mechanical properties of Hadfield
steel in its austenite form are the following: a density (q) of
7890 kg/m3, a Brinell hardness HB of 186–229, and a strength, r,
of 654–830 MPa; mechanical properties: ductile alloy. The physical
and the mechanical properties of ferro-chromium industrial-type
ores with high-melt impurities are the following: a density (q) of
2235 kg/m3, a Brinell hardness (HB) of 438–662, and a strength,
r, of 307–522 MPa. mechanical properties: fragile ore mineral.
The gating and feeding system parameter variation methods are
categorised into classic, experimental, and high-efficiency (new)
(Fig. 1).
The classic method involved a massive head for the supply of
molten metal through the gating system. After pouring the molten
metal into the mould, the form was horizontally rotated at 25
(Fig. 1a). In the experimental method, a significantly reduced head
was used in the corner of the plate. The supply of molten metal was
not through the gating system to the head; molten metal entered
from the end of the plate and there was no rotation of the form
after pouring (Fig. 1b). The new, highly efficient method permitted
the molten metal to enter from the end and the side of the plate.
The supply of molten metal through the gating system was
switched to both the end and the side of the plate. Moreover, the
form was not turned after pouring (Fig. 1b).
The tests were conducted on cast plates with the following
designs (Fig. 2a–c): a – ‘light’, b – a ‘medium’, and c – a ‘heavy’
Each plate has a matching one with negligible variation in
weight, average wall thickness, and design. These plates are widely
used in ferroalloy crushing stations, have a relatively simple
design, and their production is fraught with several thermal stress,
shrinkage, and drop defects.
The plates are conventionally classified into ‘light’, ‘medium’,
and ‘heavy’; this categorisation identifies the effect of the plate
geometrics (primarily that of the average wall thickness) on the
severity of production-related defects.
To determine the chemical composition of the alloy, spectral
analysis was performed on a modern ISKROLINE 300 static-
emission spectrometer with a concentration measuring range of
The measurement of the steel temperature as it crystallised in
the mould was performed by applying tungsten-rhenium thermo-
couples (VR 5/20) which were connected to EPR-08mz, an auto-
matic electronic potentiometer. The melt temperature was
measured in degrees Celsius.
number of inputs; the reduction of the number of inputs is an addi-
tional industrial requirement in order for such models to be imple-
mented in factories, because they will reduce the costs of analysis
Table 1
Chemical composition (%) of Hadfield steel.
(i.e. if a percentage of only 2 rather than 16 chemical components
should be evaluated in a workpiece, then the analytical process
will cost less).