Professional Documents
Culture Documents
Intelligent Engines: Revolutionizing Manufacturing and Supply Chains With AI
Intelligent Engines: Revolutionizing Manufacturing and Supply Chains With AI
A. Automation and Efficiency Machine shops and manufacturing businesses may still
Once a business decides to manufacture, its primary need to start using outdated processes and methods.
goal is to stay competitive. Growth and profit drive the However, this may not continue due to the retiring skilled
business, leading to increased pressure for better and workers and the fewer younger workers interested in the
cheaper products. Small and medium-sized businesses need field. Increased competition and material costs also make
help investing in large-scale production lines to meet current processes less effective. Intelligent software and AI
production demands. However, computer-aided can offer a solution by optimizing processes like high-speed
manufacturing and intelligent software have overcome this machine-making, reducing cycle time without requiring
issue and provide efficient solutions. These tools are crucial costly changes in the entire process. Due to limited time and
for thriving businesses and avoiding closure due to market resources, this ability to improve existing processes is highly
decline. sought after.
C. Quality Control
The second most significant cited problem was
manufacturers' inability to utilize the information available
to them to make informative decisions, with a third of
manufacturers saying data needed to be more bulky and
complex to manage. AI can help manufacturers sift through
this data to detect patterns, diagnose issues, and recommend
action.
This has the potential to improve quality, prevent D. Supply Chain Optimization
costly defects, and reduce rework, an area where Supply chains are vital but complex and fragile. AI can
manufacturers currently need help to make informed predict and solve problems, like disruptions and risks. IBM's
decisions using inductive problem-solving. 30% of AI early Watson Supply Chain and UPS's ORION platform are
adopters in manufacturing are using or planning to use AI examples. They optimize routes and offer responsive
for predictive maintenance, compared to only 6% of systems. AI can also simulate and predict cash flow impacts.
manufacturers overall, which tells of AI's potential to AmBev and AB InBev's case showed the risks of extending
improve maintenance activities and shift away from the payment terms. AI estimated a cost of US$200 million for
current cost-intensive run-to-fail maintenance strategy. AB InBev and risk to relationships with small suppliers and
farmers.
III. INTELLIGENT ENGINES: TRANSFORMING path_to_save (str): Local path where the dataset should
MANUFACTURING PROCESSES be saved.
"""
ML techniques are used in manufacturing. IBM's api = KaggleApi()
Watson, a well-known ML application, optimizes its api.authenticate()
function as a Quality Engineer. ML algorithms predict api.dataset_download_files(dataset_name,
machinery failure and reduce costs. Ecolab used ML to path=path_to_save, unzip=True)
analyze soil loading, resulting in reduced costs.
Manufacturers are increasingly using analytics and AI. def load_data(file_path):
CEOs plan to integrate cognitive computing/AI for various """
purposes in the next 3-5 years. Function to load the data from a CSV file.
Args:
Machine learning is AI that improves prediction file_path (str): Path to the CSV file.
accuracy without explicit programming. Algorithms use """
input data and statistical analysis to predict outcomes. There import pandas as pd
are three types of learning: supervised, unsupervised, and return pd.read_csv(file_path)
reinforcement.
if __name__ == '__main__':
Machine learning technologies are already being used # Define the dataset and path
effectively in manufacturing and can only become more dataset = 'username/dataset-name' # Example:
prevalent in the future. The study "The Learning Enterprise" 'username/manufacturing-engine-data'
sponsored by Accenture, and authored by Sam Khan, states save_path = './dataset'
that top executives believe machine learning can generate file_name = 'data.csv' # Adjust based on actual file
substantial value in terms of increased productivity, higher name in the dataset
efficiency, and lower operational costs.
# Download the dataset
A. Machine Learning Algorithms -Data Collection download_dataset(dataset, save_path)
Data Collection Code sudo code snippets are provided
below. # Load the data
data = load_data(os.path.join(save_path, file_name))
import requests print(data.head())
import os
from kaggle.api.kaggle_api_extended import KaggleApi # Author: Vishwanadham Mandala
def download_dataset(dataset_name, path_to_save): # Date: 2024-04-14
""" Code 1.Python data collection program
Function to download dataset using Kaggle API.
Args:
dataset_name (str): Full path of the dataset on Kaggle.
Output Snippets.
D. Robotics and Autonomous Systems where robots learn from experts or their own mistakes, is
The popularity of robotics in manufacturing is being researched. Error recovery is crucial for tasks like
increasing due to lower costs and more accessible assembly. Modular robots that can reconfigure themselves
programming. AI systems are being used to coordinate large are another option. AI techniques are used for coordination
numbers of robots, including CNC machines, for complex and planning. Robotics is expected to be a significant area
tasks like aircraft manufacturing. Inflexibility in for AI in manufacturing.
programmed paths has been a barrier, but flexible robotics,
F. Data Analytics and Decision-Making manufacturing sectors use existing data more effectively to
AI can enhance data analysis and decision-making by plan inventory and meet customer demand. Additionally,
providing superior predictive analysis tools and algorithms. AI-based tools can provide intelligence for ad-hoc analysis
These tools simplify complex analytical procedures and and simulation, enabling specific hypotheses testing and
allow all personnel to run complex analyses. AI also helps evaluating the effects of decisions.
AI allows complex decision-making in data analysis positive impact, high-level discussions should lead to new
and simulation. It can codify human expertise into a laws, company codes, and customs. As intelligent engines
decision support system, offering specific recommendations develop rapidly, regulations specific to their use are
and guiding users with no experience. AI tools store and essential. Flexibility and performance-based regulations are
reuse knowledge, combining specific intelligence about important to accommodate the challenges and changes
analysis techniques with subject knowledge. For instance, brought by intelligent engines. Preemptive health and safety
scheduling algorithms optimize task scheduling while risk assessments will ensure their beneficial introduction in
leveraging knowledge of the task or machine. the workplace.
A. Ethical Considerations
The concept of addressing the ethical aspects of AI and
robotics with robust policy responses and international
frameworks is recommended by various organizations. The
focus should shift towards the potential benefits of these
technologies rather than the perceived "evils." To ensure a
A security breach in an AI system has the potential to providing end-to-end security and safety assurance for AI
have a severe impact on society's quality of life. Measures to technologies. Additionally, it will require a shift in how
ensure the security and safety of AI systems present a society approaches security and safety as it becomes reliant
significant challenge. They will require more research into on intelligent systems.
D. Potential Applications in Other Industries company. The system was demonstrated in a single facility
In the case study, we found the successful in the manufacturing sector, specifically the supply chain for
implementation of the Intelligent Engine within one an automotive parts warehouse.
It was found that the intelligent engine improved the The intelligent engine improved the human engineers'
quality of scheduling. A team of six engineers did the schedules in only a fraction of the computational time. It is
original scheduling within the plant. Their schedules were important to note that the operation research team at the
often found to be of poor quality, and the only feasible way warehouse could have been more enthusiastic about the
to get good results was through extensive debugging within simulation result on the first day.
the simulation environment.
This was because the quality of engine schedules was use the schedules to direct the factory in a way consistent
very high, and the research team did not believe it was with the global optimality of the engine schedules. The
feasible to get such good results. However, once the engine research team relinquished their scheduling control to the
schedules were implemented, the research team found the engine, which was deemed automatic within two weeks.
quality of work to be much improved, and they were able to
IIoT improves manufacturing and supply chains, [10]. Kapoor, K., Zhou, Y., & Bhattacharya, S. (2019).
including resource utilization, waste reduction, and higher- Artificial intelligence: Transforming the nature of
quality products. Intelligent engines and AI are leading this manufacturing and the job market. Industrial
transformation by optimizing system performance, Management & Data Systems, 119(7), 1561-1587.
minimizing downtime, and automating control system https://doi.org/10.1108/IMDS-09-2018-0444
design and maintenance. AI methods support the [11]. Li, J., Tao, F., Cheng, Y., & Zhao, L. (2021). Digital
computerization of control and automate the automation twin-driven product design, manufacturing and
process. This solves the problem of knowledge acquisition service with big data. The International Journal of
for control and enables advanced control technology to Advanced Manufacturing Technology, 102(9-12),
solve complex problems. 3965-3984. https://doi.org/10.1007/s00170-019-
03637-5
REFERENCES [12]. Zhang, David Q., et al. (2008). The value of
information sharing in a two-level supply chain.
[1]. Agrawal, Ankit, et al. (2005). Collaborative International Journal of Production Economics,
forecasting process integration: An exploratory study 111(2), 298-304. [DOI: 10.1016/j.ijpe.2007.02.021]
of the current state in consumer goods supply chains. [13]. Infosys. (n.d.). Robotics & autonomous systems.
International Journal of Physical Distribution & Retrieved [formatted as April 15, 2024], from
Logistics Management, 35(6), 430-449. https://www.infosys.com/services/engineering-
[2]. Bass, Frank M., et al. (1982). Promotion planning services/service-offerings/robotics-autonomous-
models. International Series in Quantitative systems.html
Marketing, Vol. 14. New York: Springer. [14]. Subashini, P., Krishnaveni, M., Dhivyaprabha, T. T.,
[3]. Ozer, Ozalp. (2004). Inventory management when & Shanmugavalli, R. (2020). Review on Intelligent
the demand distribution is not known. Management Algorithms for Cyber Security. In P. Ganapathi & D.
Science, 50(10), 1348-1362. [DOI: Shanmugapriya (Eds.), Handbook of Research on
10.1287/mnsc.1030.0180] Machine and Deep Learning Applications for
[4]. Peterson, Leif B., et al. (2006). Order visibility in a CyberSecurity (pp. 1-22). IGI Global.
pull supply chain: Equilibrium and efficiency. Naval https://doi.org/10.4018/978-1-5225-9611-0.ch001
Research Logistics, 53, 735-752. [DOI:
10.1002/nav.20146] BIOGRAPHY
[5]. Aldoseri, A.; Al-Khalifa, K.N.; Hamouda, A.M. Re-
Thinking Data Strategy and Integration for Artificial Vishwanadham Mandala [1] is a Data Engineering Lead
Intelligence: Concepts, Opportunities, and in Data Engineering, Data Integration, and Data Science
Challenges. Appl. Sci. 2023, 13, 7082. areas. He holds bachelor’s and master’s degrees and Data
https://doi.org/10.3390/app13127082 Science master’s in computer science & engineering and
[6]. Tang, Christopher S., et al. (2002). Special issue on Data Science
supply chain visibility and performance. Decision
Sciences, 33(1), 131-142. [DOI: 10.1111/j.1540- Manogna Dolu Surabhi [2] is Quality Assurance
5963.2002.00103.x] Engineer in Ottawa Area Intermediate School District
[7]. Vishwanadham Mandala, Revolutionizing (OAISD), Holland, Michigan
Asynchronous Shipments:Integrating AI Predictive
Analytics in Automotive Supply Chains.
International Journalof Artificial Intelligence &
Machine Learning (IJAIML), 1(1), 2022, 47-59.DOI:
https://doi.org/10.17605/OSF.IO/FTXEV
[8]. Mandala, Vishwanadham, C. D. Premkumar, K.
Nivitha, and R. Satheesh Kumar. "Machine Learning
Techniques and Big Data Tools in Design and
Manufacturing." In Big Data Analytics in Smart
Manufacturing, pp. 149-169. Chapman
andHall/CRC, 2022.
[9]. V. Mandala, R.Rajavarman, C.Jamunadevi, R.Janani
and Dr.T.Avudaiappan, "Recognition of E-
Commerce through Big Data Classification and Data
Mining Techniques Involving Artificial Intelligence,"
2023 8th International Conference on
Communication and Electronics Systems (ICCES),
Coimbatore, India, 2023, pp. 720-727,
doi:https://doi.org/10.1109/ICCES57224.2023.10192
673.