ALRAQABA . ISSUE 15
79
Therefore, this thesis can be used in any type
of manufacturing regarding the size of the data.
From the first sample created, the code created
can predict the new samples with an accuracy of
about %95, with fewer defects and higher quality
with a time not exceeding 55–40 minutes (if the
data is enormous).
In auditing, especially in auditing performance, the
code created from the author could be the best
solution to monitor and predict the output needed
in the tender applied for at the State Audit Bureau
(SAB) to validate the output through the technical
specification and how they meet the requirements
submitted from the contractor. For example, if the
company (contractor) offers data about any type
of manufacturing, the auditor at the SAB can have
a %95 indication about the output of each sample
by using the code (which is user-friendly without
the need for domain experts). Therefore, the code
can easily be used in auditing performance for
any type of manufacturing. Furthermore, it can
be used to compare what the contractor submits
through the documents with the real data (only
a small sample of data is enough to predict the
rest of the required manufacture data). A detailed
summary of the thesis will be discussed below.
A Modified Random Forest algorithm (MRF)-
based predictive model is proposed for use
in manufacturing processes to estimate the
effects of several potential interventions,
such as (i) altering the operating ranges of
selected continuous process parameters within
specified tolerance limits, (ii) choosing particular
categories of discrete process parameters, or (iii)
choosing combinations of both types of process
parameters. The model introduces a non-linear
approach to defining the most critical process
inputs by scoring the contribution made by each
process input to the process output prediction
power. It uses this contribution to discover optimal
operating ranges for the continuous process
parameters and/or optimal categories for discrete
process parameters. The set of values used for
the process inputs was generated from operating
ranges identified using a novel Decision Path
Search (DPS) algorithm and Bootstrap sampling.
Image A
The odds ratio is the ratio between the
occurrence probabilities of desired and undesired
process output values. The effect of potential
interventions, or of proposed confirmation trials,
are quantified as posterior odds and used to
calculate conditional probability distributions. The
advantages of this approach are discussed in
comparison to fitting these probability distributions
to Bayesian Networks (BN).
Image B
The proposed explainable data-driven predictive
model is scalable to many process factors with
non-linear dependence on one or more process
responses. It allows the discovery of data-driven
process improvement opportunities that involve
minimal interaction with domain expertise. An
iterative Random Forest algorithm is proposed to
predict the missing values for the mixed dataset
(continuous and categorical process parameters).
It is shown that the algorithm is robust even at high
proportions of missing values in the dataset.
Image C
The number of observations available in
manufacturing process datasets is generally low,
e.g. of a similar order of magnitude to the number
of process parameters. Hence, Neural Network
(NN)-based deep learning methods are generally
not applicable, as these techniques require (50
100-) times more observations than input factors
(process parameters).
The results are verified on several benchmark
examples with datasets published in the literature.
The results demonstrate that the proposed
method outperforms the comparison approaches
in term of accuracy and causality, with linearity
assumed. Furthermore, the computational
cost is both far better and very feasible for
heterogeneous datasets.
Image D and Image E
Theses
