Computational Management Science (2023) 20:14

https://doi.org/10.1007/s10287-023-00448-0

ORIGINAL PAPER

Using machine learning prediction models for quality

control: a case study from the automotive industry

Mohamed Kais Msakni1 · Anders Risan1 · Peter Schütz1

Received: 3 June 2022 / Accepted: 1 March 2023 / Published online: 16 March 2023
© The Author(s) 2023

Abstract
This paper studies a prediction problem using time series data and machine learning
algorithms. The case study is related to the quality control of bumper beams in the
automotive industry. These parts are milled during the production process, and the
locations of the milled holes are subject to strict tolerance limits. Machine learning
models are used to predict the location of milled holes in the next beam. By doing
so, tolerance violations are detected at an early stage, and the production flow can be
improved. A standard neural network, a long short term memory network (LSTM),
and random forest algorithms are implemented and trained with historical data,
including a time series of previous product measurements. Experiments indicate
that all models have similar predictive capabilities with a slight dominance for the
LSTM and random forest. The results show that some holes can be predicted with
good quality, and the predictions can be used to improve the quality control process.
However, other holes show poor results and support the claim that real data prob-
lems are challenged by inappropriate information or a lack of relevant information.

Keywords Neural network · Random forest · Quality control · Manufacturing

Mathematics Subject Classification 68T07 · 68T09 · 62H10

Anders Risan and Peter Schütz contributed equally to this work.

* Mohamed Kais Msakni

kais.msakni@ntnu.no
Anders Risan
andris@stud.ntnu.no
Peter Schütz
peter.schutz@ntnu.no
1
Department of Industrial Economics and Technology Management, Norwegian University
of Science and Technology, Torgarden, 7491 Trondheim, Norway

13
Vol.:(0123456789)
14 Page 2 of 28 M. K. Msakni et al.

1 Introduction

The emergence of the fourth industrial revolution, Industry 4.0, is primarily driven
by advancements in information, communication, and intelligence technologies that
can improve production flexibility, efficiency, and productivity in industry (Ibarra
et al. 2018). While the definition of Industry 4.0 is broad, there are several key con-
cepts associated with it, such as smart factories, the Internet of Things (IoT), cloud
computing, cyber-physical systems, and Big Data Manufacturing (Santos et al.
2017). IoT technology enables to connect manufacturing resources, like sensors,
machines, and other equipment, enabling interconnection between components and
reducing human intervention. This also allows real-time, high-accuracy monitoring
of product quality, equipment, and production processes. Real-time data flow can
help identify problems early on and provide better visibility into the flow of materi-
als and products. In addition, cloud computing makes data available to other systems
with powerful resources, such as servers, storage, and software (Lee and Lee 2015).
As many manufacturers have large amounts of data that go unused, cloud comput-
ing is seen as a way to transform the traditional manufacturing business model into
an effective collaboration, helping manufacturers align business strategies and prod-
uct innovation and create smart networks (Xu 2012). The amount of data collected
from various systems and objects is growing at an exponential rate and is commonly
referred to as Big Data. This concept is characterized by high dimensionality and
high complexity due to the variety of formats, semantics, and quality of sensors and
processes generating the data (Wuest et al. 2016). As a key concept in smart facto-
ries, Big Data can impact Industry 4.0 in three ways: enabling self-diagnosis, fore-
casting, and control (Tao et al. 2017). Conventional data processing software and
technologies cannot fully leverage the potential of these large and complex datasets,
and advanced methods such as machine learning algorithms are needed to organize
and derive value from the data.
In the context of Industry 4.0, machine learning has been applied to different lev-
els of the industrial process, such as anomaly detection, process optimization, pre-
dictive maintenance, quality control, diagnosis, and resource management (Roblek
et al. 2016). Machine learning is seen as a promising improvement in manufacturing
as it allows for decentralized, autonomous, and real-time decision-making without
human interaction. It has the advantages of addressing large and complex processes
and enabling continuous quality improvement (Dogan and Birant 2021). Unlike con-
ventional algorithms, machine learning algorithms can dynamically learn from the
system and automatically adapt to changes in the environment. It can also detect
patterns and implicit knowledge from the data, improving existing processes and
methods in manufacturing (Wuest et al. 2016). However, the application of machine
learning is not straightforward. The performance of these algorithms can be hindered
by the acquisition of relevant data in terms of volume and quality. On the one hand,
the training data must be sufficiently numerous to reach the level of generalization,
for which the learning model also performs well on new (unseen) data. On the other
hand, the data may either contain inappropriate and redundant information or lack
relevant information, as not all data is captured during the manufacturing process,

13
Using machine learning prediction models for quality control:… Page 3 of 28 14

and some attributes may not be available. Data preprocessing, which includes select-
ing relevant inputs and normalizing the data (Wuest et al. 2016), is also an important
step before learning. The challenges of machine learning are not only limited to data
but also include the algorithm itself. Some machine learning algorithms are more
appropriate for specific applications, and the performance of some of them depends
on selecting suitable hyperparameter settings. Despite these challenges, machine
learning algorithms have the capacity to extract new information and provide better
results than conventional algorithms.
One of the advances offered by Industry 4.0 is the opportunity to improve quality
control in manufacturing. Traditionally, manufacturers have used Statistical Process
Control (SPC) to ensure that product features are defect-free and meet specifications.
SPC is based on the statistical assumption that random factors, such as humidity,
temperature changes, and variations in raw material, tend to form a normal distribu-
tion centered on the quality characteristics of the product (e.g., length, weight, and
hardness). Thus, the process is under statistical control, which allows for analyzing
the outputs and the capability of the process. SPC provides tools and techniques for
monitoring and exploring the process behavior and identifying anomalies (Tao et al.
2017; Oakland and Oakland 2018). With the technological capabilities of Industry
4.0, SPC can be supplemented to improve quality control further. Big data and cloud
computing can use real-time data to detect quality defects and process instability at
an early stage. For example, Gokalp et al. (2017) describe real-time data analysis to
self-calibrate a process when a deviation in the trajectory of an ongoing machining
process. In addition, machine learning can use time-series data of process and prod-
uct variables to identify patterns and detect early process deviations so that preven-
tive measures can be taken and the production process is stabilized.
This paper investigates the use of machine learning algorithms to predict product
quality in manufacturing in order to support quality control. The focus is on bumper
beams, which are an essential component of automotive crash management systems
and are subject to strict quality control. The goal is to improve the quality control
process in production by predicting the quality of future products, allowing for early
adjustments, and reducing scrap production and downtime in the production system.
The machine learning algorithms used in this study are based on neural networks
and random forests. They are trained on historical data consisting of previously pro-
duced and measured parts provided by the manufacturer. The effectiveness of the
neural network and random forest models is compared and evaluated for their ability
to predict key product characteristics important for quality control. This work differs
from previous research in that it develops machine learning models that use previ-
ously measured products to predict the quality of the next product rather than using
the real-time state of the system to predict the quality of the current part.
The outline of the remainder of this paper is as follows. Section 2 discusses
machine learning for quality control in manufacturing systems and presents related
works in the literature. Section 3 introduces the concept of time series and relates it
to process control and machine learning prediction models. The case study of this
paper is discussed in Sect. 4. Section 5 shows the implementation and the obtained
performance of the learning models. Finally, Sect. 6 gives a conclusion to this paper.

13
14 Page 4 of 28 M. K. Msakni et al.

2 Related works

Maintaining high-quality products and processes is essential for success in a

competitive environment. In manufacturing, product quality is related to the func-
tional aspects of the product that must be out of defects and out-of-tolerance con-
ditions. The process of ensuring that any manufactured product meets the require-
ments is called quality control. If a product does not satisfy the requirements, it is
considered a poor-quality product and will be removed from the production line.
Many factors can cause quality to vary in the production process, such as humid-
ity, temperature, and variations in raw materials and tools.
As technology advances and data becomes more available, new ways to per-
form more accurate, real-time quality control are emerging. Machine learning has
already been successfully applied to tasks involving quality control and quality
assessments and is expected to further improve the field of quality control in the
future (Wuest et al. 2016). Collected data can be analyzed by learning algorithms
in two ways (Tao et al. 2018). The first is to monitor the process in real-time to
ensure product quality; for example, a deviation in tool trajectory can be detected
using real-time analysis, and the process can be adjusted according to the require-
ments. The second is to identify emerging problems. Using historical data, the
learning algorithms can identify patterns or predict the output characteristics of a
process, enabling early detection of faulty products.
The neural network is one of the most widely used machine learning algo-
rithms for process and quality control in manufacturing environments. Most of
the applications are related to real-time analysis for process control or detection
of defective products by image recognition. Karayel (2009) uses a feedforward
neural network as an observer for a control system by predicting surface rough-
ness in a computer numerical control (CNC) lathe. The prediction model uses
process parameters as input data, i.e., cutting depth, cutting speed, and feed rate,
to predict surface roughness. This prediction is later sent to the controller to
determine the best cutting parameters for the CNC turning system. A similar net-
work structure is used by Tsai et al. (1999) to predict real-time surface roughness
in milling cutting operations. The model uses process parameters that consist of
vibration measures, depth of cut, speed of the axis, and rotation and feed speed.
Martin et al. (2007) propose a supervised feedforward neural network to replace
a human expert in the quality control process of resistance spot welding. The
machine learning model uses ultrasonic oscillograms to classify the quality spot
welds into one of the six predefined levels. Zhao et al. (2020) use power signals
to predict the nugget diameter from spot-welded joints in a real-time prediction
system. The paper compares the performance of a regression model and feedfor-
ward network in monitoring weld quality and shows that the latter model provides
better performance. To detect defects in parts, Wang et al. (2018) propose a deep
convolutional neural network that uses raw images of flat surfaces to automati-
cally extract product features for defect detection and improve production effi-
ciency. For geometrically complex products (such as turbo blades), Wang et al.
(2020) develop a similar model architecture in a cloud-based platform to meet the

13
Using machine learning prediction models for quality control:… Page 5 of 28 14

high-speed performance required by complex product images. Risan et al. (2021)

develop a feedforward neural network to predict the location of a milled hole in a
bumper beam.
Other works in the literature focus on prediction models to support quality control
using random forest algorithms. With respect to the machining process, most works
use process parameters as input variables for prediction models such as feed rates,
tool wear, and drive power. Bustillo et al. (2021) investigate the prediction of surface
flatness deviations in a face milling process using different machine learning algo-
rithms. The input data consists of tool life and wear and drive power. The problem
is first designed and evaluated as a regression problem and then as a classification
problem using discretized flatness levels. For the regression problem, the random
forest is outperformed by the two artificial neural networks, a multilayer percep-
trons network and a radial basis functions network. However, when the classifica-
tion problem is considered, the random forest gives the most accurate predictions.
Wu et al. (2018) develop a random forest model to predict the surface roughness in
fused deposition modeling. The input data is mainly based on the temperature and
vibration of the table and extruder. The model can predict the surface roughness
of a printed part with very high accuracy. Bustillo et al. (2018) propose different
machine learning algorithms for surface roughness and loss-of-mass predictions in
machining processes. The models studied include regression trees, multilayer per-
ceptrons, radio basis networks, and random forest. The experiments show that mul-
tilayer perceptrons achieve the best surface roughness prediction. However, the ran-
dom forest has the advantage of being more suitable for industrial use in the absence
of experts in machine learning as this model has a non-parametric property. Agrawal
et al. (2015) develop a multiple regression model and a random forest model for the
prediction of surface roughness during hard turning of a hardened steel piece. Both
models use the same cutting parameters as input variables. The results show better
surface predictions for the random forest model. Li et al. (2019) use an ensemble
of machine learning models to predict the surface roughness for an extrusion-based
additive manufacturing process. A random forest is used to reduce the input variable
size to improve the computational efficiency and avoid overfitting. Then, an ensem-
ble of different machine learning models is used for surface roughness prediction.
Quality control can also be enhanced using time series data and prediction mod-
els. Ma et al. (2022) proposed a soft sensor model for quality prediction of industrial
products using time series data and process features. The model framework is based on
a neighborhood dimension reduction and a bidirectional gated recurrent unit. Shohan
et al. (2022) use time series modeling to help improve the prediction quality of bio-
fabrication process. Standard autoregressive time series models and machine learning
models were tested. Experiments showed that the Long-Short Term Memory model
provides the best performance in terms of mean square errors. In another applica-
tion, Freeman et al. (2018) implement deep learning techniques in the prediction of
air quality. The data consists of time series events of hourly air quality and meteoro-
logical events. Kim et al. (2022) use a descriptive time series analysis to predict down-
time of a linear medical accelerator by using long-term maintenance data. Meng et al.
(2022) deploy a deep learning model using time series data of historical images and
image recognition of plants. The objective is to improve the quality by helping have

13
14 Page 6 of 28 M. K. Msakni et al.

healthy plants with high yields. Prediction using time series data is not limited to qual-
ity improvement but covers a wide range of applications. Many works have recently
emerged to predict COVID-19 transmission using time series and deep learning mod-
els (Long Short Term Memory networks and Gated Recurrent Units), e.g., Rauf et al.
(2021); Ayoobi et al. (2021).

3 Methodology

Time series data, which consists of observations recorded at specific times, is often
available in manufacturing processes and equipment. It is then important to exploit
these data to extract valuable information for the manufacturers. This task corresponds
to finding a model that describes a time series. This model estimates the relationship
between the variable of interest Y and the input variables X using a function f. While
various approaches can be applied, i.e., physical, statistical, and machine learning mod-
els, the nonlinear and high-dimensional aspects of manufacturing systems make it very
difficult to develop a satisfactory model for estimating f. Despite this challenge, devel-
oping a time series model has several advantages, such as a compact description of the
time series, hypothesis testing, separation and filtering of noise from data, and time
series prediction (Brockwell and Davis 2016).

3.1 Times series and statistical process control

In the context of quality control, SPC is a widely used method that involves process
capability and statistical analysis of process results. These methods rely on monitoring
and analyzing the product features relevant to product quality. By using samples of a
specific size from the process, causes for variation can be identified, and adjustments
can be made (Groover 2019).
One of the primary techniques is the control chart, which offers a visualization way
to study the evolution of a process over time. A time-series data is represented in a
chart with a central line for the average, an upper line for the upper control limit, and
a lower line for the lower control limit. These control limits are then compared to the
actual data to see if the process variation is under control. When the process is under
statistical control, the control limits are defined based on the process capability (PC),
which provides information about the accuracy of a process’s performance over time
and measures the ability of a process to meet its specifications (Oakland and Oakland
2018). It can be defined as:
PC = 𝜇 ± 3𝜎 (1)
where 𝜇 is the mean of the process, and 𝜎 is the standard deviation. Thus, 99.73% of
outputs of a controlled process are within 3𝜎 limits.

13
Using machine learning prediction models for quality control:… Page 7 of 28 14

3.2 Machine learning prediction models

The main challenge of machine learning models is to establish a valid representation

of the input data (the input variables X of the time series) by performing some trans-
formation (the model or function f) that can approximate the expected outcomes (the
variable of interest Y). The provided data set is commonly referred to as the “train-
ing set" and is used by machine learning algorithms to apply a predefined set of
operations to build a model. Thus, machine learning models are not explicitly pro-
grammed to make decisions or predictions but are created during the learning stage.
Machine learning has been successfully applied to a variety of problems across dif-
ferent domains, such as image recognition, anomaly detection, and quality control.
In this work, two classes of machine learning algorithms are developed, namely neu-
ral networks and random forests, to predict the location of holes in future products.
Since the locations are continuous values and the training set is composed of input
and output variables, the problem is referred to as a regression problem with super-
vised learning. A general description of neural networks and random forests is given
in the following subsections.

3.2.1 Neural networks

Neural networks are one of the most commonly known machine learning algo-
rithms, and they have been successfully applied to a wide range of fields. This learn-
ing algorithm is inspired by a biological network of neurons, in which neurons are
chemically connected to form an extensive network. In artificial neural networks,
neurons are modeled as nodes and connections as weights. The role of weights is
to computationally activate or deactivate a connection between two nodes: A posi-
tive weight indicates an active connection, whereas a negative weight prohibits the
link between the nodes. A node receives many connections (weights) that are trans-
formed into a single output. Typically, the neurons in a neural network are organized
in layers. The first (input) layer passes the input data to the network without any
transformation, and the last (output) layer consists of output variables. The hidden
layers connect the input layer to the output layer and perform the data transformation
using activation functions. The role of an activation function in the hidden layers is
to transform the weighted sum of the input into an output that will be used in the fol-
lowing layers. A general structure of the feedforward network is illustrated in Fig. 1.
The layered representation of neurons can represent complex relationships
between input and output data and extract complex patterns. Indeed, neural networks
can model non-linear-statistical data and handle high-dimensional and multivari-
ate data. However, they require more data than other machine learning models, and
the best performance requires extensive customization as neural networks depend
on several hyper-parameters. Also, neural networks do not provide any information
about how the outputs are computed, a problem commonly referred to as the black
box in machine learning (Ian et al. 2016).
Although neural networks are well suited for a large variety of problems, such
as image recognition and text recognition, they suffer from a major issue known as
the vanishing gradient problem, which prevents learning long-term dependencies

13
14 Page 8 of 28 M. K. Msakni et al.

Fig. 1  An example of a feedforward neural network with an input layer, two hidden layers, and one out-
put layer with one target variable. Adapted from (Ketkar and Moolayil 2021)

(Rehmer and Kroll 2020). This makes it difficult to train standard neural networks
on long data series (Kinyua and Jouandeau 2021). The vanishing gradient problem
can be addressed by including gated units, such as the Long Short-Term Memory
and the Gated Recurrent Unit (Rehmer and Kroll 2020).

3.2.2 Random forests

The random forest algorithm has become a widely used machine learning algorithm
because of its simplicity and accuracy (Biau and Scornet 2016) and its ability to per-
form both supervised and unsupervised learning, as well as classification and regres-
sion problems (Genuer and Poggi 2020). This algorithm is a statistical learning
method proposed by (Breiman 2001), based on the principles of ensemble learning.
In machine learning, ensemble learning refers to the techniques of combining the
predictions of a group of trained models (an ensemble). The idea is that by aggregat-
ing the outcomes of several models, the prediction of the ensemble is more likely to
perform better than any individual model in the ensemble. For the random forest,
the algorithm is trained on different and independent training subsets (bootstraps) to
obtain several models, referred to as trees. Figure 2 illustrates a general structure of
the random forest.
A decision tree is a predictive model with a tree-like structure where the deci-
sion progresses from the root node through internal nodes until it reaches a leaf. A
node corresponds to a binary split of the predictor space to continue the decision
flow in one of the two sub-trees of the node. A leaf in the decision tree represents
a predicted value or class label, and the path to the leaf represents the classification
rules. Such a decision representation makes decision trees readable and simple to
interpret. Although there are different algorithms for building a decision tree, the

13
Using machine learning prediction models for quality control:… Page 9 of 28 14

Fig. 2  Flowchart of training a random forest tree and aggregation of results. Adapted from (Genuer and
Poggi 2020)

classification and Regression Tree (CART) algorithm is widely used for random for-
ests (James et al. 2013).
For a classification problem, the decision tree uses the input values to reach one
of its leaves and, here, to find the predicted class. Similarly, a regression problem
uses the same decision tree structure, with the difference that the leaves correspond
to continuous target values. To make the decision trees independent of each other,
they are trained on B randomly drawn and independent subsets of equal size. Each
subset is used to train B decision trees that will finally be aggregated into a forest.
One advantage of the random forest algorithm is that it does not require heavy
computation for training. It is easy to tune as it depends only on a few hyper-param-
eters. Another advantage is that it is suitable for high-dimensional problems with
multivariate data where the number of variables far exceeds the number of observa-
tions, and vice versa (Géron 2019). However, the prediction quality of the random
forest is highly dependent on the quality of the training set, e.g., it cannot predict
values outside the minimum and maximum of the values in the training set.

4 A case study

The product studied in this paper is the bumper beam, a component of a crash man-
agement system in cars. The beam is formed from an extruded aluminum profile and
is machined and cut before being assembled to the bumper with screws.
The beam is placed using clamps at predefined locations during the machining
step. Then, the CNC machining starts with the milling of reference holes, which are
of particular interest because they are used to locate and mill other holes by CNC

13
14 Page 10 of 28 M. K. Msakni et al.

machining. In total, there are 20 milled holes, each with narrow tolerance ranges
regarding their location in the beam. Any displacement of the reference holes results
in deviation of the connected holes. Quality control of the milled holes is performed
after the machining process, when a new product is released, or at predefined inter-
vals. The interval length between two quality controls is typically two hours and is
considered by the manufacturer as satisfactory to guarantee high-quality standards
while ensuring smooth production. During quality control, the geometric character-
istics of all milled holes and the beam curvature are automatically measured in an
XYZ grid system, resulting in a total of 144 different features. When the measure-
ment report shows any deviation, the entire batch of products is rejected, and the
production line is disrupted. The production batch since the last control is scrapped.
An experienced operator makes the necessary changes to the machine settings. Then
a new beam is machined, and another quality control is performed. The goal of the
manufacturer is to reduce the downtime of the production as much as possible to
minimize direct economic loss.
Many factors can cause variations in the CNC machining process, including both
random variations such as clamping force, temperature, and variations in upstream
activity, as well as assignable variations such as replacement of CNC parts and
change in the beam type being processed. Unfortunately, not all of these variations
are available to be considered as part of the input to the learning models.
Figure 3 illustrates the shape of the bumper beam and the locations of the ref-
erence holes. Two reference holes (H1 and H4) are located on the left side of the
beam, and three other holes (H2, H3, and H5) are located on the right side of the
beam. H1, H2, and H3 are located using the YZ coordinate system, whereas H4 and
H5 are located using the XZ coordinate system. This work aims to improve the qual-
ity control of the milled holes by predicting the reference hole locations of the next
product to be manufactured. Machine learning models are implemented to predict
future hole positions, which can be used as a preventive measure to avoid out-of-tol-
erance products. Since this information is available, early adjustments can be made,
and the production flow is smooth. The proposed learning models do not depend
on real-time data, as is the case in many literature works, but consider a time series
analysis that uses previous measures to predict the hole locations in the upcoming
product. Historical data from all available measurements is used as input to train the
models. The target variables are the coordinates of all reference holes.

Fig. 3  Illustration of the bumper beam shape and the locations of the five reference holes

13
Using machine learning prediction models for quality control:… Page 11 of 28 14

Fig. 4  A table from a control report of the studied beam showing the measurements of the reference hole
H1

Figure 4 shows an example of the measurements and data collected for the refer-
ence hole H1. This hole is located using a measured value (MS) and a nominal value
(NM). The deviation (DV) of H1 is the difference between MS and NM. Based on
the deviations of Y and Z, denoted here by dy and dz, the true position (TP) of the
measured hole can be computed. The TP defines a circular tolerance area for the
hole position and is defined by Eq. 2.
√
TP = 2 dy2 + dz2 (2)

For the example of Fig. 4, the TP measure is within the predefined tolerances limits
(−T and +T ). The angular deviation (DA) complements the TP measure and pro-
vides information about what direction the hole has moved. The actual location of
H1 related to this example is represented by a dotted circle in Fig. 5.
The TP values of the other holes are calculated similarly using the deviations
of the two coordinates locating a hole. The decision as to whether a hole location
is within specification or not depends on the TP values, which must be within the
lower and upper limits defined by the manufacturer.

5 Experiments and results

In this section, the data, implementation, and performance of each machine learning
model are discussed.

Fig. 5  An illustration of the

displacement of H1 of Fig. 4.
The related milled hole is rep-
resented by a dotted circle. The
solid circle corresponds to the
nominal position, and the dotted
square represents the area where
the hole meets the specifica-
tions. DA shows the angular
deviation of the milled hole

13
14 Page 12 of 28 M. K. Msakni et al.

5.1 Training, validation and test set

The dataset used for the quality control prediction consists of 1255 measurement
reports, covering three years. Each report includes a timestamp of the measurement
operation, the locations of the 20 milled holes, and the curvature of the beam, result-
ing in 144 different point measurements. It should be mentioned that the interval
between two quality control measurements is not always two hours and can vary
greatly depending on the production schedule, holidays, priorities, etc. As shown
in Fig. 6, which depicts the measurements of two hole-coordinate pairs using a
time-stamped axis, the production for the bumper beam under study was partially
interrupted during November and December 2019. This kind of interruption can be
found several times (about 15 times) throughout the dataset, with most of them last-
ing for one or two weeks. Despite these interruptions, we assume that the dataset is
continuous as the number of interruptions is small, and the machine learning models
used in this study depend only on lag features for prediction.
For a given learning algorithm, the variable to be predicted (the output) is a sin-
gle hole-coordinate pair, e.g., H1-Y, that is trained and tested separately. The predic-
tion of measure t uses all points of the three lagged measurements, i.e., t − 3, t − 2,
and t − 1, as input, resulting in 3 × 144 independent variables for every variable to
predict. Indeed, preliminary testing showed that the prediction mainly depends on
the last observation t − 1, and that it can be slightly improved by integrating three-
lagged input. Furthermore, it should be stated that all measures are related to rela-
tive deviation. These values are available with the raw data used in this study.
The dataset is divided into two subsets. The first 70% of the dataset is used to
train and validate the machine learning algorithms, and the remaining 30% is used
to test the prediction performance of the models. Since the default hyperparame-
ters of all models cannot guarantee optimal results for the prediction problem, we

Fig. 6  Scatter plot of the measurement values of H1-Y and H2-Z over the data collection period

13
Using machine learning prediction models for quality control:… Page 13 of 28 14

performed a hyperparameter tuning step that was done on the first subset of data.
This subset was, in turn, divided into two other subsets, where the 50% of the data-
set is used for training, and the next 20% is used for validation. The hyperparam-
eters test was done on a randomly chosen hole, namely H5-X (the results of subsec-
tion 5.4 show that this hole has an average performance). The best parameters found
were then used for the other holes. It should be noted that other holes were selected
for the hyperparameter tests, and similar results were obtained.

5.2 Implementation

In addition to the random forest, two neural network models are considered for
prediction purposes. The first model is based on a standard neural network, here-
after referred to simply as a ‘neural network’, and the second is a Long-Short Term
Memory (LSTM). All machine learning models were implemented in Python 3.8.6
using the Scikit-Learn library (for the neural network and the random forest) and the
Keras library (for the LSTM). The input data is processed using Pandas 1.1.3 and
Numpy 1.19.2, and the visualization tools are based on Matplotlib 3.3.2. The work-
ing environment is Jupyter Notebook on a Windows machine with a Core i7 CPU
and 32 GB of RAM.
The parameters of the neural network are one hidden layer of 100 neurons, and
the LBFGS solver is used as the identity activation function. The hyperparameter
function of the random forest is used to find the best parameters, with the best results
obtained with the default settings. A greedy hyperparameter tuning was performed
for LSTM, where the number of epochs ranged between 1 and 1000, the batch size
was set to 1, 2, and 4, and the number of neurons was set to 1, 2, 4, and 10. The
best parameters were found for 100 as the number of epochs, two as the number of
batches, and one as the number of neurons.
Finally, it should be mentioned that the neural network and LSTM models are
more sensitive to data scaling than the random forest. The input data is pre-pro-
cessed by standardizing, and the same scaling is then applied to the input of the test
set.

5.3 Data analysis

The first step in analyzing the collected data is to understand the problem and verify
its quality. This step involves visualizing and evaluating the relevance of the data,
identifying outliers, and removing bad entries. Figure 6 presents the different meas-
ured values for the hole-coordinate pairs H1-Y and H2-Z over the data collection
period. It can be observed that the H2-Z measurements are more spread out than
H1-Y measurements. In contrast, H1-Y measurements fall within a narrow range
that varies over time without a distinctive trend (e.g., degradation over time). These
variations could potentially be explained by changes in the production process, but
without additional data, it cannot be confirmed. Furthermore, the dispersion of the
values in Fig. 6 (especially for H1-Y) supports the idea of using lagged measure-
ments for prediction purposes.

13
14 Page 14 of 28 M. K. Msakni et al.

(a) Box-and-whisker plots for H1-Y, H1- (b) A distribution plot for H1-Y, H2-Z,
Z, H2-Z, H4-Y, and H5-Y. and H5-Y.

Fig. 7  Analysis of measurement data for a subset of reference holes and coordinates

Figure 7a illustrates the distribution of the different measurements of hole-coor-

dinate pairs. Since showing all pairs in the same figure makes it difficult to read, we
restrict the representation to only five variables that are selected randomly and have
data properties compared to other variables. Figure 7a indicates that all variables
share similar data distribution properties. The median is almost equal to zero, and
the interquartile ranges (IQRs) presented by boxes are very narrow, revealing that
the measurements are concentrated within this area, especially for H1-Y, H1-Z, and
H4-Y. In addition, Fig. 7a shows for each pair the outliers, defined as the points out-
side of the whiskers, at 1.5 of IQR from the first and third quartiles. We can see that
the majority of outliers are condensed at the head and tail of whiskers, and very few
measurements are far from the rest of the data. Thus, the data available for this study
is considered to be of good quality and does not require preprocessing.
Furthermore, Fig. 7b illustrates the distribution plots of pairs of hole-coordinate.
For the sake of readability, we select only three representative pairs as they have
similar distributions to the other holes. The H2-Z plot shows a normal distribution
that is symmetric and bell-shaped but slightly deviated from the zero mean. That
is, this particular pair of hole-coordinate can be represented by a standard normal
distribution. As for the H1-Y measurements, the distribution has a slight multimodal
shape that can be smoothed to a normal distribution with high kurtosis, meaning
that most observations have zero deviation. However, H5-Y has different peaks with
different densities, which means that most measurements are slightly deviated from
zero.

5.4 Performance of the models

In this subsection, the performance of the neural network, LSTM and random forest
are assessed from both a quantitative and qualitative perspective. In addition, the
models are compared to a standard autoregressive time series model.

13
Using machine learning prediction models for quality control:… Page 15 of 28 14

5.4.1 A quantitative comparison

Three common performance metrics for regression problems are used to evalu-
ate the predictive quality of models. The first is the Mean Absolute Error (MAE)
which shows the magnitude of the overall error between the observed and predicted
values. It neither eliminates the effect of positive and negative errors nor penalizes
extreme forecast errors. The second is the Mean Squared Error (MSE), which penal-
izes extreme values. A high value of MSE shows a significant deviation between
observed and predicted values, whereas a low value indicates that the predicted val-
ues are very close to the observations. Finally, the root mean square error (RMSE) is
commonly used for regression problems and measures the square root of the second
sample moment of residuals. RMSE is used to compare prediction errors of different
models for the same data set and a particular variable, as it is scale-dependent. The
definition of these three metrics are given in Eqs. (3), (4), and (5).
n
1�
MAE = ‖y − ŷ i ‖ (3)
n t=1 i

n
1∑
MSE = (y − ŷ i )2 (4)
n t=1 i

√
√ n
√1 ∑
RMSE = √ (y − ŷ i )2 (5)
n t=1 i

where,

yi is the observed target value,

ŷ i represents the predicted target value, and
n is the number of observations.

Figure 8 shows the MAE, MSE, and RMSE metrics for the Random Forest and
the two versions of Neural Network. Except for the reference hole H2, all models
can provide reasonable predictions relative to the actual observations, i.e., the MAE
metric ranges from 0.11 to 0.28 mm for the other holes. In particular, the predic-
tions for hole H1 provide the best metrics, and H2 has the worst prediction metrics
for the Y and Z coordinates. When learning models are considered, LSTM provides
the best performance for H2-Z, where the metric MAE is improved by 31% over the
neural network. For the remaining hole-coordinate pairs, the average MAE is the
same for all models, i.e., 0.16. However, Random Forest and LSTM perform slightly
better than Neural Network for MSE, i.e., 0.045 against 0.047.
Figure 9 groups the hole-coordinate pairs that are located in the same direction
and provides the MAE and MSE metrics of each direction by learning model.
Together with the results shown in Fig. 8, we observe that for holes in the YZ
coordinates (H1, H2, and H3), all prediction errors in the Z coordinate are higher

13
14 Page 16 of 28 M. K. Msakni et al.

Fig. 8  Comparison between the Random Forest, Neural Network and LSTM models using the MAE,
MSE and RMSE metrics for the locations of all reference holes

Fig. 9  Comparison between the prediction models by grouping the holes located in the same coordinate
(X, Y or Z). The metrics are MAE and MSE

than the corresponding Y coordinate. The same pattern appears for holes in the
XY coordinates (H4 and H5) where the prediction errors in the X direction are
slightly higher than in the Y direction. Since the comparison by coordinate is
consistent for all learning models, it can be concluded that the models are better
suited to one direction than another. Furthermore, when considering the locations
of the holes in the beam, it can be observed that the holes on the left side of the
beam, i.e., H1 and H4, are better predicted than the holes on the right side of the
beam, i.e., H2, H3, and H5.

13
Using machine learning prediction models for quality control:… Page 17 of 28 14

The previous analysis is further extended to include the feature importance of the
predicted hole-coordinate pairs in the Y direction. The metrics of H2-Z and H3-Z
show poor predictions compared to H1-Z, despite all of them being located in the
same direction. Therefore, it is worth exploring which variables are the most signifi-
cant and which factors have the greatest impact on the prediction. This can be done
by analyzing the average feature importance of the decision trees in the random for-
est model. Figure 10 shows the top 20 most important features for H1-Z, H2-Z, and
H3-Z. It can be observed that there is a variety of variables involved in the impor-
tance feature, including the previous measurements of the hole to be predicted, with
the lagged time in parenthesis, as well as bend and other (non-reference) hole meas-
urements. In the label in front of a variable name, the letter T refers to the twist
tolerance for a bend measurement. The same letter is also used for small holes and
indicates the distance to a reference hole (a specific metric set by the manufacturer is
used). Lastly, the DF denotes the diameter of a milled hole.
Figure 10 shows that three factors affect the prediction quality. First, all predicted
pairs depend strongly on their immediate previous measurement t − 1 and, to a lesser
extent, on t − 2 and t − 3. However, basing the predictions solely on previous meas-
urements leads to poor prediction metrics, as can be seen in H2-Z. Second, a diverse
range of information leads to better predictions. The feature importance of H1-Z
shows that different sources of information are used, i.e., Bend 66 and Bend 67 are
bend measurements near the location of H1, Bend 61 is in the middle of the beam,
and Bend 20 and Bend 21 are on the other side of the beam. Third, having only

Fig. 10  Top 20 most important features of the random forest model for the hole-coordinate pairs in the Y
direction (H1-Z, H2-Z and H3-Z)

13
14 Page 18 of 28 M. K. Msakni et al.

low-importance values does not help to have good predictions (i.e., H3-Z). This
indicates that the learning model cannot identify the relevant features for a good
prediction of the target variable. Overall, this analysis confirms the importance of
considering all available information and three-lagged measurements for prediction
purposes.

5.4.2 A qualitative comparison

Figure 11 gives a qualitative comparison of the prediction models for the best-per-
forming hole-coordinate pair, namely H1-Y. The bottom part of Fig. 11 plots the
actual values and those predicted by the random forest, while the top part draws the
predictions of Neural Network and LSTM (since they belong to the same family of
learning models) together with the actual values. It can be observed that, in general,
the random forest provides restricted and smooth predicted values. This observation
is notable for the first and last segments of observations, where the values predicted
by the random forest are always within the fluctuations of the actual measurements.
However, LSTM and neural network models can track the spikes better to generate
predictions as high as the actual values. Except for the measurements around Obser-
vation 100, LSTM performs marginally better than the neural network. Overall, all

Fig. 11  Prediction performance of the neural network (top) and random forest (bottom) models for the
best-predicted coordinate – H1-Y

13
Using machine learning prediction models for quality control:… Page 19 of 28 14

Fig. 12  Prediction performance of the neural network (top) and random forest (bottom) models for the
worst predicted coordinate – H2-Z

H1-Y predictions can be considered of high quality with good performance for both
the random forest and LSTM.
In the second qualitative comparison, Fig. 12 illustrates the performance of the
prediction models for the worst-performing hole-coordinate pair, namely H2-Z. It
can be seen that there is a significant gap between the actual observations and the
predicted values for all models. Except for the first 40 observations, the random for-
est and neural network models generate poor predictions compared to the actual val-
ues. For the interval between observations 40 and 270, the neural network attempts
to follow the trend of the actual values without providing good predictions, while the
random forest predicts deviations close to zero. However, the LSTM shows much
better performance as the actual values are better tracked. Indeed, when this segment
of observations is considered, the MAE of LSTM is 0.24 against 0.36 and 0.30 for
the neural network and random forest, respectively. This explains the better metrics
obtained by LSTM for H2-Z. From about observation 270, the deviation of the pre-
dicted values from the actual observations becomes increasingly significant for all
models. In particular, the random forest generates a prediction close to zero. We can
conclude that this last segment is very peculiar; unknown changes have been made
to the production process preventing the learning models from making good predic-
tions. The limited performance is not due to limited learning capacity but rather to
missing information not provided to the models. Indeed, as previously discussed and

13
14 Page 20 of 28 M. K. Msakni et al.

shown in Fig. 8 and Fig. 9, the prediction is better in one direction than in another.
Similarly, the prediction of the holes located on the left side of the beam is better
than the holes located on the other side of the beam. This may be due to the clamp-
ing forces applied to the beam during the machining process, which is not available
for this study. Another reason may be a variation in the upstream activity, for exam-
ple, when the aluminum profiles are bent.

5.4.3 Comparison with an autoregressive time series model

The performance of the machine learning models is compared to an autoregressive

integrated moving average (ARIMA) model. The model was fitted to the time series
data to predict future points in the series. We recall that ARIMA is a univariate
model, which means that previous data of a specific hole-coordinate pair is used to
predict the next observations. The ‘statsmodels’ library was used to implement the
ARIMA model, and the pre-built ‘auto_arima’ function was called to identify the
most optimal values for trend elements (p, d, q), where p, d, and q are the trends in
autoregression order, difference order, and moving average order, respectively. The
best parameters differ from one hole to another, for example, the ARIMA (3,1,3)
model is used for H1-Y, and the ARIMA (2,0,4) is used for H2-Z.
The metrics of the ARIMA models are shown in Fig. 13 and compared to the ran-
dom forest and LSTM models. To compute the MAE and MSE, a static prediction
for ARIMA for all pairs is used, meaning that the fitted models employ the actual
value of the lagged dependent variable for prediction. It is clear that the learning
models perform much better than the ARIMA models for all holes. This indicates
that integrating other information such as bend and other holes’ measures help to
achieve a better prediction. The exception is for H2-Z where ARIMA performs bet-
ter than the random forest and is comparable to LSTM. This can be explained by
the fact that ARIMA provides a very restrictive prediction that is always within the
bounds of the actual values, which helps to reduce the error, especially for the last
observation segment for H2-Z. However, as shown in Fig. 12, the random forest and
LSTM perform poorly for this last segment, starting at observation 270. Despite
this poor performance, the learning models provide important information for this

Fig. 13  A comparison between ARIMA, random forest, and LSTM using MAE and MSE metrics

13
Using machine learning prediction models for quality control:… Page 21 of 28 14

particular segment. They indicate that unknown changes have been made to the pro-
duction process. This information cannot be derived from the ARIMA results.

5.5 Residual evaluation

The prediction quality of the proposed models can be further assessed by analyz-
ing the residuals, which are the difference between the actual and predicted values.
These residuals should be uncorrelated and normally distributed with zero mean
(Kuhn and Johnson 2013). Figure 14 shows the histograms of residuals for the hole-
coordinate pairs studied above, namely H2-Z and H1-Y, for all models. In addition
to the worst- and best-predicted coordinates, the analysis includes H5-X and H4-X,
which are ranked around 7th and 3rd positions in terms of MAE for all models
considered.
The histograms in Fig. 14 provide insight into the distribution of residual errors
and the prediction quality. In particular, the distribution of residuals for H2-Z (the
worst-predicted coordinate) is non-Gaussian and positively skewed with high kur-
tosis (large tails) for the neural network and random forest. This confirms the poor
prediction quality for these two prediction models. However, LSTM shows a better

Fig. 14  Residual normality test for the predictions of LSTM, Neural Network, and Random Forest for the
worst and best MAE metric, H2-Z and H1-Y, respectively. The comparison also includes H5-X and H4-X
ranked 7th and 3rd on the same MAE metric

13
14 Page 22 of 28 M. K. Msakni et al.

distribution of residuals for H2-Z, which confirms the better performance obtained
for this coordinate compared to the other two prediction models. For the H1-Y coor-
dinate, the distribution of residuals shows a mean close to zero and a low kurtosis.
The histograms confirm the strong performance obtained with the H1-Y coordinate.
As for H5-X and H4-X, the visualization of the residuals is close to the Gaussian
distribution, especially for H5-X with the neural network. With the mean of the
distribution for H5-X and H4-X being almost equal to zero for all models, it can
be concluded that the predicted and actual values are not correlated. The predic-
tion performance of the learning models is generally good for some hole-coordinate
pairs.

5.6 Confidence bounds and TP‑outliers detection

Figure 15 shows the predicted TP values for the hole H1 (discussed in Sect. 4),
along with the corresponding 3𝜎 level (shown with orange dashed line) and the TP
limits set by the manufacturer (between 0.0 and 1.0). The predicted TP value is cal-
culated according to Eq. 2 and is based on the coordinates H1-Y and H1-Z predicted
by the random forest model. This figure aims to assess whether the predicted TP
values are within statistical control and whether they can be used for quality control
of the bumper beam. The validation test is used for the 3𝜎 control. As can be seen,
the TP limits are stricter than the 3𝜎 level. Only two outliers out of four are above
the confidence bound. When the remaining holes are considered, a similar observa-
tion is obtained, i.e., the TP limits and 3𝜎 are almost the same. The only exception is
for the hole H4, for which the 3𝜎 level is stricter than the TP limit; however, this can
be considered acceptable as only one observation of H4 deviates from both TP and
3𝜎 limits. In conclusion, the process variation is under control and subject to random
factors.
In the last experiment, Figs. 16, 17, 18, 19 and 20 compare the real and pre-
dicted TP values for holes H1–H5. The TP limits of each hole, as set by the manu-
facturer, are highlighted with a dashed green line. These figures show a difference
between the actual and predicted TP values. The results of the prediction model

Fig. 15  Predicted TP values for the hole H1 using the random forest model with a 3𝜎 level. The TP limits
for H1 are between 0.0 and 1.0 as set by the manufacturer

13
Using machine learning prediction models for quality control:… Page 23 of 28 14

Fig. 16  Comparison between the predicted and true TP values for the hole H1. The random forest model
is used to predict the coordinates of H1

Fig. 17  Comparison between the predicted and true TP values for the hole H2. The random forest model
is used to predict the coordinates of H2

Fig. 18  Comparison between the predicted and true TP values for the hole H3. The random forest model
is used to predict the coordinates of H3

13
14 Page 24 of 28 M. K. Msakni et al.

Fig. 19  Comparison between the predicted and true TP values for the hole H4. The random forest model
is used to predict the coordinates of H4

Fig. 20  Comparison between the predicted and true TP values for the hole H5. The random forest model
is used to predict the coordinates of H5

cannot follow the fluctuations of the actual TP values. However, when it comes to
outliers, the model can give some insights into when the deviations might occur.
For the hole with the best-predicted coordinates (H1, shown in Fig. 16), the pre-
dicted TP values are located in the same observation area where actual TP devia-
tions are observed. The first area, located around observations 210-213, is detected
in advance by the prediction model before observation 210. The second area records
consecutive actual TP measures that exceed the upper limit. Although the prediction
model cannot detect all of these outliers, it has a prediction located in this area. The
last area includes only observation 344, which is perfectly detected by the prediction
model with a similar actual TP value.
The same statement is also valid for H3, with the deviations of observations 226
and 272 being correctly reported by the model. Although the deviation of observa-
tion 45 is not detected, the predicted value of TP is very close to the upper limit,
which may indicate that an early adjustment of the CNC machining settings should
be made. As for H4, only one deviation is reported that is perfectly predicted by
the model. While the results are satisfactory for the holes discussed above, the

13
Using machine learning prediction models for quality control:… Page 25 of 28 14

prediction of TP outliers for H5 is poor. The predictions of the X and Y coordinates

of H5 are among the worst, which may explain the poor quality of the prediction of
TP values for this hole. Finally, no conclusions can be drawn from the analysis of
the H2 results because there are no outliers for this set of observations. However, it
is unlikely that the prediction model would be able to identify outliers for H2, as the
H2-Y and H2-Z coordinates are poorly predicted.
Furthermore, the accuracy of the learning model is evaluated using the True
Positive Rate (TPR or Sensitivity) and the True Negative Rate (TNR or Specific-
ity) measures. The accuracy is defined as the ability of the learning model to pre-
dict actual TP-outliers. In this case, measuring the specificity is not relevant as the
number of actual TP values within limits is significantly higher than actual outliers.
Therefore, the focus is on the ability of the model to correctly predict actual TP
outliers. As the model can detect outliers in nearby areas, the definition of a True
Positive is expanded to include predicted outliers that will actually occur within the
next 24 h. A False Positive is defined as when no actual TP outlier occurs during the
next 24 h, and a False Negative is defined when there is an actual TP-outlier that is
not predicted within the previous 24 h. Equations (6), (7), and (8) define TPR, False
Negative Rate (FNR), and Threat Score (TS), respectively.
True Positives
TPR =
True Positives + False Negatives (6)

False Negatives
FNR = (7)
False Negatives + True Positives

True Positives
TS =
True Positives + False Negatives + Fale Positives (8)

Table 1 reports the number of outliers, TPR, FNR, and TS rates for each hole. The
performance of holes H4 and H5, while noteworthy, is not significant since they are
only related to a single outlier that the model either predicts well or badly. For hole
H1, Table 1 indicates low TPR and TS rates, which is a result of some predictions
and actual TP outliers being separated by more than 24 h. For instance, the predicted
TP outlier at observation 200 is separated by more than 24 h from the actual TP out-
liers at observations 213-215, and similarly with the predicted TP outlier at obser-
vation 272 and the actual outliers at subsequent observations. In the case of hole

Table 1  Accuracy of the Hole #. Outliers TPR (%) FNR (%) TS

learning model for detecting
TP outliers within a 24-hour H1 12 30 70 0.27
interval
H2 0 0 0 0.00
H3 4 50 50 0.50
H4 1 100 0 1.00
H5 1 0 100 0.0

13
14 Page 26 of 28 M. K. Msakni et al.

H3, the predicted TP outlier at observation 226 is correctly reported by the learning
model; however, the next actual outlier at observation 232 occurs after 24 h, explain-
ing the relatively low TS score obtained for H3. Overall, Table 1 confirms that the
learning model cannot accurately predict TP outliers. Nonetheless, the predicted
information can still be used by the manufacturer to make early adjustments.

6 Conclusion

This paper deals with a prediction problem for quality control. The underlying prob-
lem is related to the automotive industry, and the product under study is the bumper
beam, subject to stringent quality criteria. To support the quality control process of
this product, we proposed machine learning models to predict the location of the
reference holes of the next produced beam. The models are based on a time series
that consisting of the historical data set of previous measurements that includes the
beam characteristics. The learning models developed are a neural network, a long
short-term memory network, and a random forest, and all are trained under simi-
lar conditions. The experimental study showed that the performance of all models
is generally quite similar, with a slight dominance of the long short-term memory
network and the random forest models. The results also indicate that the prediction
can be good for some hole-coordinate pairs. However, there are considerable dis-
crepancies for some other coordinates, and the predictions deviate significantly from
the actual values. Since both models showed similar behavior, it can be concluded
that the available information is not sufficient for prediction and that other resources
should be included, such as process parameters or data from an upstream activity.
This work shows that applying machine learning models to real-life problems is
not as easy as it sounds and is hampered by several factors. Not all data is captured
or made available to be used for other purposes. For example, in the context of this
work, information about changes in CNC settings is volatile and cannot be retrieved
later, limiting its use for learning purposes. This example also shows that the tran-
sition to Industry 4.0 is not a straightforward process and could be challenging in
several areas.

Funding Open access funding provided by NTNU Norwegian University of Science and Technology
(incl St. Olavs Hospital - Trondheim University Hospital). This work was supported by the Research
Council of Norway as part of the LeanDigital research project, number 295145.

Declarations
Conflict of interest One of the co-authors of this manuscript is a member of the editorial board of Com-
putational Management Science. The authors declare no-conflict of interest regarding the publication of
this paper.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as
you give appropriate credit to the original author(s) and the source, provide a link to the Creative Com-
mons licence, and indicate if changes were made. The images or other third party material in this article
are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the

13
Using machine learning prediction models for quality control:… Page 27 of 28 14

material. If material is not included in the article’s Creative Commons licence and your intended use is
not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission
directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​
ses/​by/4.​0/.

References
Agrawal A, Goel S, Rashid WB et al (2015) Prediction of surface roughness during hard turning of AISI
4340 steel (69 HRC). Appl Soft Comput 30:279–286
Ayoobi N, Sharifrazi D, Alizadehsani R et al (2021) Time series forecasting of new cases and new
deaths rate for covid-19 using deep learning methods. Results Phys 27(104):495
Biau G, Scornet E (2016) A random forest guided tour. Test 25(2):197–227
Breiman L (2001) Random forests. Mach Learn 45(1):5–32
Brockwell PJ, Davis RA (2016) Introduction to time series and forecasting. Springer texts in statistics.
Springer, Cham, pp 73–96
Bustillo A, Pimenov DY, Matuszewski M et al (2018) Using artificial intelligence models for the
prediction of surface wear based on surface isotropy levels. Robot Comput Integr Manufact
53:215–227
Bustillo A, Pimenov DY, Mia M et al (2021) Machine-learning for automatic prediction of flatness
deviation considering the wear of the face mill teeth. J Intell Manufact 32(3):895–912
Dogan A, Birant D (2021) Machine learning and data mining in manufacturing. Expert Syst Appl
166(114):060
Freeman BS, Taylor G, Gharabaghi B et al (2018) Forecasting air quality time series using deep learn-
ing. J Air & Waste Manag Assoc 68(8):866–886
Genuer R, Poggi JM (2020) Random forests. Springer, London
Géron A (2019) Hands-on machine learning with scikit-learn, keras and tensorflow: concepts, tools,
and techniques to build intelligent systems. O’Reilly Media
Gokalp MO, Kayabay K, Akyol MA, et al (2017) Big data for industry 4.0: a conceptual framework.
In: proceedings - 2016 international conference on computational science and computational
intelligence, CSCI 2016 pp 431–434
Groover MP (2019) Fundamentals of modern manufacturing: materials, processes, and systems, vol
7. Wiley
Ian G, Yoshua B, Aaron C (2016) Deep learning. The MIT Press, Adaptive Computation and Machine
Learning
Ibarra D, Ganzarain J, Igartua JI (2018) Business model innovation through Industry 4.0: a review.
Procedia Manufact 22:4–10
James G, Witten D, Hastie T et al (2013) An introduction to statistical learning, vol 112. Springer
Karayel D (2009) Prediction and control of surface roughness in CNC lathe using artificial neural net-
work. J Mater Process Technol 209(7):3125–3137
Ketkar N, Moolayil J (2021) Feed-forward neural networks. Deep learning with python pp 93–131
Kim KH, Sohn MJ, Lee S et al (2022) Descriptive time series analysis for downtime prediction using
the maintenance data of a medical linear accelerator. Appl Sci 12(11):5431
Kinyua P, Jouandeau N (2021) Sample-label view transfer active learning for time series classifica-
tion. In: international conference on artificial neural networks, Springer, London pp 600–611
Kuhn M, Johnson K (2013) Applied predictive modeling, vol 26. Springer, New York
Lee I, Lee K (2015) The internet of things (IoT): applications, investments, and challenges for enter-
prises. Bus Horiz 58(4):431–440
Li Z, Zhang Z, Shi J et al (2019) Prediction of surface roughness in extrusion-based additive manufac-
turing with machine learning. Robot Comput Integr Manufact 57:488–495
Ma L, Wang M, Peng K (2022) A novel bidirectional gated recurrent unit-based soft sensor modeling
framework for quality prediction in manufacturing processes. IEEE Sens J 22(19):18,610-18,619
Martin O, Lopez M, Martin F (2007) Artificial neural networks for quality control by ultrasonic test-
ing in resistance spot welding. J Mater Process Technol 183(2–3):226–233
Meng Y, Xu M, Yoon S et al (2022) Flexible and high quality plant growth prediction with limited
data. Front Plant Sci 13:304–989

13
14 Page 28 of 28 M. K. Msakni et al.

Oakland RJ, Oakland JS (2018) Statistical process control. Routledge

Rauf HT, Lali M, Khan MA et al (2021) Time series forecasting of covid-19 transmission in asia
pacific countries using deep neural networks. Pers Ubiquitous Comput 20:1–18
Rehmer A, Kroll A (2020) On the vanishing and exploding gradient problem in gated recurrent units.
IFAC-PapersOnLine 53(2):1243–1248
Risan A, Msakni MK, Schütz P (2021) A Neural Network Model for Quality Prediction in the Auto-
motive Industry. In: IFIP advances in information and communication technology. Springer: Ber-
lin pp 567–575
Roblek V, Meško M, Krapež A (2016) A complex view of industry 4.0. SAGE Open 6(2):1–11
Santos C, Mehrsai A, Barros AC et al (2017) Towards industry 4.0: an overview of European strategic
roadmaps. Procedia Manufact 13:972–979
Shohan S, Hasan M, Starly B et al (2022) Investigating autoregressive and machine learning-based time
series modeling with dielectric spectroscopy for predicting quality of biofabricated constructs. Man-
ufact Lett 33:902–908
Tao F, Cheng J, Qi Q et al (2017) (2017) Digital twin-driven product design, manufacturing and service
with big data. Int J Adv Manufact Technol 94(9):3563–3576
Tao F, Qi Q, Liu A et al (2018) Data-driven smart manufacturing. J Manufact Syst 48(C):157–169
Tsai YH, Chen JC, Lou SJ (1999) An in-process surface recognition system based on neural networks in
end milling cutting operations. Int J Mach Tools Manufact 39(4):583–605
Wang T, Chen Y, Qiao M et al (2018) A fast and robust convolutional neural network-based defect detec-
tion model in product quality control. Int J Adv Manufact Technol 94(9–12):3465–3471
Wang Y, Liu M, Zheng P et al (2020) A smart surface inspection system using faster R-CNN in cloud-
edge computing environment. Adv Eng Informat 43(101):037
Wu D, Wei Y, Terpenny J, (2018) Surface roughness prediction in additive manufacturing using machine
learning. ASME, (2018) 13th international manufacturing science and engineering conference.
MSEC 2018:3
Wuest T, Weimer D, Irgens C et al (2016) Machine learning in manufacturing: advantages, challenges,
and applications. Prod Manufact Res 4(1):23–45
Xu X (2012) From cloud computing to cloud manufacturing. Robot Comput Integr Manufact 28(1):75–86
Zhao D, Wang Y, Liang D et al (2020) Performances of regression model and artificial neural network in
monitoring welding quality based on power signal. J Mater Res Technol 9(2):1231–1240

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.

13