Literature Review for Research Paper: AI Manufacturing

AI Manufacturing
Name:
Institution:
Course code:
Date:
Introduction
Artificial intelligence is tipped to have a massive influence on marketing strategies and customer behaviors in the future. At the moment, AI has ceased to be the only hype as more managers and institutions are implementing it in the supply chain. Manufacturers use the supply chain to manage goods, information, money, and interaction with the consumer. The AI comes into play due to machines and systems' ability to display some aspects of human intelligence. Therefore, it is easier to automate business processes that allow AI algorithms to perform well-defined tasks requiring little to no human intervention. These tasks can range from recordkeeping, sorting out emails, managing language in systems, and implementing simple economic transactions. Consequently, AI implementation ranges from power shifts, reassignment of decision-making responsibilities, enhancing effectiveness, downsizing, and cost reduction. This essay is a literature review that seeks to analyze the impacts on profitability to organizations that use AI compared to those that don't.
Technology trends have provided evidence that the convenience of all processes can be achieved with enough innovation. AI's use in project management is basically to help integrated systems perform tasks without human input. The consensus is that AI delivers higher quality and efficient outcomes than human experts (Anderson & Raine, 2018). It is, therefore, safe to assume that AI and machine learning are going to replace humans. While this is a complicated topic, the benefits brought forth by AI cannot be underestimated. For instance, AI is very efficient in improving the supply chain related to efficiencies in customer services. Customer interaction is always at the forefront of a manufacturing organization's strategy; therefore, AI's effectiveness helps satisfy the company while also saving money CITATION Pet93 \l 1033 (Duchessi, O'Keefe, & O'Leary, 1993).
Organizations that use AI achieve the benefits of reducing inventory, labor, and time, resulting in productivity over time (Toorajipour et al., 2021). For effective processes, large manufacturing firms rely heavily on the accurate distribution of information and data. Transfer of data is critical to the manufacturing process as it allows the supply chain to continue. For instance, organizations like the automobile company are original manufacturing industries. Therefore, to make products for the end consumer, information needs to be effectively transferred within the company and other organizations along the supply chain. The AI is responsible for informing the supplier of how many cars are in production and what order; consequently, the supplier delivers the exact number of parts needed to complete it. Much of this process is data-driven; therefore, AI helps manage to track the inventory, a process that can be overwhelming for humans to do. Consequently, the AI helps in time management by improving efficiency and quickly spotting deficiencies in the production process.
Theories
AI in itself is complicated due to its ability to reason and act like humans. Therefore, with each distinctive feature, AI can be classified into various sub-fields. They include artificial neural networks (ANN), rough set theory, machine learning, expert systems, GAs, and agent-based systems CITATION Min10 \l 1033 (Min, 2010). Therefore, an organization's profitability can vary depending on which type of AI sub-fields a company chooses to prioritize. Artificial neural networks are based on how human brain cells and neurons conduct their functions. Therefore, ANN can learn through experience, recognize patterns, process abstract information, and cluster objects together (Abiodun et al., 2019). For instance, a land vehicle manages to stay in a single lane by mimicking a human driver's actions. Such technology is expensive to implement, and companies that don't have it can compete. However, when it is fully integrated into manufacturing processes and products, profits will increase significantly.
Rough set theory is capable of synthesizing the approximation of concepts and acquired data based on classification attributes. Introduced by Pawlak (1982), the theory can help an organization select the best supplier from a pool of qualified suppliers by using conflicting supplier selection criteria. Machine learning is a theory by Samuel (1995) designed to investigate computers' ability to acquire knowledge from data without being programmed. Machine learning is based on learning tasks and mimic human actions through neurological programming (Calegari et al., 2020). Therefore, it can be a valuable tool in assessing data to determine the motivations behind the increase or decrease in collaboration levels in an organization.
On the other hand, expert systems are based on the same criteria of computers capable of emulating human skills and cognitive abilities. The application of these systems is limitless in supply as it has shown tremendous effectiveness in air traffic control, spatial mapping, airline yield management, and vehicle repair and maintenance schedule (Cheung, Ip, & Lu, 2005). Apart from sophisticated forecasting, expert systems can help in contemporary product design and planning, influencing an organization's cost-effectiveness. Genetic algorithms work by imitating natural evolution rules, selecting processes, and fitting into a surrounding environment. The algorithms have been used to solve combinational problems by encoding possible solutions to numerical strings called chromosomes. Another problem-solving system is the agent-based system that distributes the main problem into sub-problems capable of being assessed by independent entities. An agent represents an autonomous entity capable of taking specific actions to accomplish a particular set of goals (Jennings, Sycara, & Wooldridge, 1998). An agent's characteristics are its ability to exploit a massive amount of domain knowledge and learn from the decision environment.
Methodology.
This literature reviews studies adopted through two types of research methods; qualitative and quantitative. Qualitative research relies on expressing opinions through words, while quantitative research relies on numerical data to conduct mathematical analysis. By use of qualitative methods, the motivations and opinions on the interpretation of AI can be achieved. Consequently, it helps develop a hypothesis and collect data needed for the quantitative study. The data collection involves using primary data, a collection of information utilized for the first time. Secondary data; information already used previously by other researchers is also utilized CITATION Mar19 \l 1033 (Munir, 2019). Secondary data is based on the analysis of opinions, thoughts, and ideas from different authors to collect more opinions. For this, journal articles and books are used to determine the impacts of AI on profits.
To manufacturing companies, time is the equivalent of money. Therefore, AI offers considerable advantages to operations compared to companies that still rely on traditional automation. For instance, the average Fortune 100 company aims to shorten the supply chain to one day, which helps in freeing up US$50 million to US$100 million in cash flow. Therefore, it makes sense that companies are heavily investing in AI to provide the organization with effective supply chain automation. Companies such as Tesla and Johnson & Johnson have business models that rely on cohesive globally integrated automation. Therefore, Elementum, an AI start-up, is responsible for maintaining a streamlined supply chain.
The profitability of AI has seen companies like Elementum venture into the world of AI to provide the service to some of the world's biggest manufacturers. Services offered range from tracking transportation, monitoring one-off incidents, recording manufacturing output, and providing real-time supply visibility. In a day, the company can analyze more than 10 million incidents in the company and close to US$25 trillion worth of products, all in real-time (Purdy, M., & Daughterty, 2017). Consequently, the manufacturers can receive an early warning of potential problems in the supply chain and propose solutions. The efficiency AI provides can be seen in an incident involving a Chinese DRAM chip factory in 2014. Fire managed to halt almost 25% of the supply, inconveniencing equipment manufacturers that knew about the incident days later (Garside, 2013). However, Elemenutm customers got the information only minutes after the incident and managed to secure their DRAM supply before the shortage affected the prices. Besides production and supply chain...