Article
Developing a Recommendation Model for the Smart
Factory System
Chun-Yang Chang
1
, Chun-Ai Tu
2,
* and Wei-Luen Huang
3
Citation: Chang, C.-Y.; Tu, C.-A.;
Huang, W.-L. Developing a
Recommendation Model for the
Smart Factory System. Appl. Sci. 2021,
11, 8606. https://doi.org/10.3390/
app11188606
Academic Editor: João Carlos de
Oliveira Matias
Received: 3 July 2021
Accepted: 13 September 2021
Published: 16 September 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 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/).
1
Department of Intelligent Commerce, National Kaohsiung University of Science and Technology,
Kaohsiung 807618, Taiwan; cyc@nkust.edu.tw
2
Business Intelligence School, National Kaohsiung University of Science and Technology,
Kaohsiung 807618, Taiwan
3
Kaohsiung Veterans General Hospital, Kaohsiung 807618, Taiwan; wlhuang@vghks.gov.tw
* Correspondence: 1103405108@nkust.edu.tw
Abstract:
In Industry 4.0, the concept of a Smart Factory heralds a new phase in manufacturing; the
Smart Factory System (SFS) will have a huge demand in Taiwan. However, the cost of constructing
a factory system will be high, and the complexity processes and introduction time must be consid-
ered. Thus, it is important to figure out how to grasp the key success factors for Smart Factories to
reduce difficulties in the process, deal with the occurrence of problems, and improve the success
rate of constructing Smart Factories. This research constructs an SFS recommendation model to
make up for past research deficiencies in terms of recommendation. It combines the methodology
of the Engel–Kollat–Blackwell Model (EKB Model) and the Modified Delphi Method to derive SFS
recommendation indicators. Through analyzing weights, the ELECTRE II was used to obtain the
importance of each dimension by calculating the Modified Compound Advantage Matrix. For proto-
type indicators, it reviewed the past literature to find out deficiencies and examined the world’s four
largest manufactories or computer technology corporations to analyze their Smart Factory solutions
regarding the SFS function characteristics. The survey ran for several rounds with a group of five
experts to amend indicators until a consensus was obtained. It proposed 64 indicators of 8 primary
dimensions in total, based on the Updated Information System Success Model, and then added the
concepts of SFS Function characteristics, Information Security, Perceived Value, Perceived Risk, and
UI Design. According to the indicators, the framework and prototype of this system will provide
solutions and references for purchasing SFS, the functions of which include SFS purchase ability
analysis, demand analysis of manufacture problems, and raking and scoring of recommendation
indicators. It will provide real-time ranking and the best alternative recommendations to suppliers,
and will not only be referred to for design and modification but also enable the requirements to be
closer to the users’ demands.
Keywords:
Smart Factory; recommendation model; recommendation system; Modified Delphi
Method; ELECTRE II Method
1. Introduction
The Smart Factory has become quite important in the manufacturing field in recent
years, as big data, cloud computing, and the Internet of Things (IoTs) have changed the way
of manufacture, leading to Industry 4.0 [
1
–
5
]. The market value of Smart Factory was at
120.98 billion USD in 2016 and is expected to grow approximately 9.3% to 188.72 billion USD
between 2017 and 2022. (Markets and Markets, 2017). Grand View Research estimated that
the global smart manufacturing market size will reach 395.2 billion USD by 2025 (Grand
View Research, 2017). From the above, it is evident that the Smart Factory is quite important.
The Smart Factory System (SFS) is a solution that combines big data, cloud computing,
and IoTs to drive manufacture automation, leading to better production efficiency [
6
–
9
].
In practical application, SFS is clouded, synchronizing data in real-time. Combined with
Appl. Sci. 2021, 11, 8606. https://doi.org/10.3390/app11188606 https://www.mdpi.com/journal/applsci
