Citation: Bomba, G.; Ornat, A.;
Gierlak, P.; Muszy´nska, M.
On-Machine Measurements for
Aircraft Gearbox Machining Process
Assisted by Adaptive Neuro-Fuzzy
Inference System. Appl. Sci. 2022, 12,
4780. https://doi.org/10.3390/
app12094780
Academic Editors: Fang Cheng,
Tegoeh Tjahjowidodo, Qian Wang
and Ziran Chen
Received: 12 April 2022
Accepted: 8 May 2022
Published: 9 May 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
On-Machine Measurements for Aircraft Gearbox Machining
Process Assisted by Adaptive Neuro-Fuzzy Inference System
Grzegorz Bomba
1
, Artur Ornat
1
, Piotr Gierlak
2,
* and Magdalena Muszy ´nska
2
1
Pratt & Whitney Rzeszów S.A., ul. Hetma´nska 120, 35-078 Rzeszów, Poland;
grzegorz.bomba@prattwhitney.com (G.B.); artur.ornat@prattwhitney.com (A.O.)
2
Department of Applied Mechanics and Robotics, Faculty of Mechanical Engineering and Aeronautics,
Rzeszow University of Technology, al. Powsta ´nców Warszawy 12, 35-959 Rzeszów, Poland;
magdaw@prz.edu.pl
* Correspondence: pgierlak@prz.edu.pl
Featured Application: The results of the work may find potential application in the final quality
control of parts on a production line.
Abstract:
This paper deals with the development of dimensional control technology for the pro-
duction of accessory drive train (ADT) gearbox housing, according to the closed door technology
approach. The work presents the methodology of the final inspection of bearing seat position de-
viation by replacing the coordinate measuring machines (CMMs) with a computerized numerical
control (CNC) machine and adaptive neuro-fuzzy inference system. The results of the work indicated
that correct solutions were obtained. In addition, the technological process of manufacturing is fully
automated and performed entirely on the production line.
Keywords:
closed door technology; accessory gearbox; on-machine measurements; CNC; CMM; ANFIS
1. Introduction
The subject of this paper includes quality control of aircraft casing covers for acces-
sory drive train (ADT) gearboxes produced by way of the casting machining process.
Due to the responsibility of these systems, all produced pieces are subject to inspection.
A typical production quality control usually consists of in-process inspection of selected
dimensional and shape-related characteristics by operators and in final inspection in a
measurement laboratory using coordinate measuring machines (CMMs) [
1
]. The control
includes, without limitation, measurement of parallelism and perpendicularity of surfaces,
measurement of surface flatness, and measurement of dimensions of holes and their true
position. The advantages of CMMs are widely known and are related mainly to high
precision and capability to measure in an automatic manner, as well as the generation of
measurement reports. The biggest disadvantage of their usage is the high cost of purchase
and maintenance. Therefore, measurement laboratories are very often the bottleneck of
the production system [
2
,
3
]. This is due to the fact that a control cell equipped with a
CMM usually supports multiple production lines. This also raises the problem of storage
of manufactured products prior to measurement, as well as the involvement of technical
resources and human resources in the in-plant transportation of products.
Modern production systems in the aviation industry, and not only in this industry,
are designed so that the human factor has the lowest possible impact on the quality of
products [
4
]. This simultaneously enhances occupational health and safety [
5
]. It entails,
however, the necessity to automatize as many operations as possible. Fully automated
production systems realize the so-called closed door technology (CDT) [
6
]. CDT is charac-
terized by maximizing machine functionality through proper task planning and eliminating
human influence on the process [
7
,
8
]. One of the ways in which this is implemented is the
Appl. Sci. 2022, 12, 4780. https://doi.org/10.3390/app12094780 https://www.mdpi.com/journal/applsci
