LOSS PREVENTION IN BUSINESS USING ARTIFICIAL INTELLIGENCE: A CASE STUDY OF AMAZON . Business loses if you spend more money than you earn. Businesses often work short-term losses at the beginning or at growth times. As this is a report so I will try to follow the report format. You can download the complete report at the bottom of the blog.
Aims: This study hopes to bring light on the use of emerging technology, Artificial intelligence, in business to prevent loss.
Objective: A survey was conducted in an institute of Islamabad with the help of a standard questionnaire regards to knowledge of AI, how AI could be used to prevent business loss.
Findings: The survey showed astounding results of 65% of the respondents already knowing about this problem and the other 45% had little knowledge about it.
Conclusion:With the help of surveys and cited references, it was concluded that many people are aware of the use of AI in business and where it is most prominent. However, they don’t have complete knowledge of how AI can prevent loss.
Recommendation: A suggestion was put forth to hold more seminars and events about technology so people could get an idea of AI and its uses in business.Amazon should use AI because there is less chances of human error when AI systems are used.
Business loses if you spend more money than you earn. Businesses often work short-term losses at the beginning or at growth times. This should not be a problem if someone has enough money in the bank to cover the cost of running your business until your income is a certain profit. There are many business failures related to document reading, processing, analysis, and data capture. In addition, employees repeat the same tasks. Finally, in most cases, the data is completed in person. For example, an employee who copies numbers fills in customer surnames in a spreadsheet, calculates financial results, and so on. It is a digital age and such inconsistencies in business processes are a different matter. However, when someone has to double-click the same area on the screen, use Ctrl + C and Ctrl + V or press the marker in the database many times or thousands of times, it is not surprising that by 1001, someone is there. may choose the wrong option. With some setbacks, such as working from home the risk increases.
There are many things that cause mistakes in the workplace. Things like lack of experience, or loss of attention to fatigue may cause employees to forget to complete tasks, to send documents to co-workers or the wrong customer, or to type errors that affect contract terms. These mistakes may seem small but such problems have caused businesses to lose millions every year. Some people worry that AI can eventually lead to joblessness, but the general thinking is there will be a need for job creation and new roles emerging to facilitate the transition to this new environment.
1.2- Purpose of report
The purpose of this report is to solve problems in business loss because of biodiversity with the help of Artificial Intelligence. With the use of technology, there are fewer chances of error and calculation can be easily done in seconds. Repetitive tasks are better fulfilled by technology. When industries become automated, but businesses have just started using software robots.
1.3- Significance of the Study
We see a growing number of robotic process automation (RPA), process and content intelligence projects in the U.S. and globally. Everyone needs their processes to be faster and less costly, intelligent automation with Artificial Intelligence is a good option.
1.4- Review of Related Literature
Fraud and spam Detection:
Amazon has Built models that help to estimate potentially corrupt transactions or identify unsuitable item reviews. This new feature helps to reduce online frauds by highlighting suspicious online payment transactions before processing order procedure.AI helps to distinguish accurately between legal and risky accounts registrations and one can selectively introduce additional checks like two-factor authentication or email verification.
Customer relationship management:
AI is enterprising Amazon customers with promotions and customer service awareness. The models are trained based on the AI algorithm of telecom companies, which gives statistics about the number of customers quitting their landline business for online competitors. The data is then used by Amazon analysts to find out which customers have a higher chance of pulling out of the company.
Amazon Content Personalization mechanism:
Amazon allows developers to build applications with the AI technology used by Amazon.com which can deliver a wide range of personalization occurrences. Amazon is an entirely organized machine learning service that goes with sharp static rules-based recommendation trains to deliver highly demanded suggestions to customers across industries such as retail, media, and entertainment.
Amazon involves customized content suggestions related to products you want to buy, entertainment you might like. previously, these systems were powered by humans. However, with the advancement of artificial intelligence, algorithms can now identify the items that might interest the customer by scanning their previous purchases and item visits and comparing that information accordingly.
Speech Recognition with Amazon:
Alexa and various other assistants detect the words and return feedback. Computers have recently been able to listen and respond to people which is Amazon’s biggest success. Diverse accents and different speech patterns of humans make it a complex task, hence consuming more traditional math and computer science. With AI and ML, the organization of algorithms can more easily understand what was spoken.
1.5-Methods of Study
1.5.1- Source of Data
The questionnaire was designed by a group of students. Each question had different answers related to the nature of the question. The respondents were advised to give their opinion on emerging field AI and how it has influenced the business administrative overall roles in the world.
1.5.2- Sample Selection
The questionnaire was distributed among 50 students. To ensure confidentiality, the questionnaires did not contain personal data forms. However, the possibility of duplication was erased by taking respective measures.
1.5.3- Statistical Methods
According to the questionnaire, a percentage of the answers was calculated to get a better understanding of the results. All the questions were evaluated because of the response they received. The percentage was calculated for each question by comparing the answers and dividing them by the total number of answers.
1.1- Scope of Study
The study includes the results that have been concluded by conducting research through issuing questionnaires in Islamabad. Therefore, the nature of the sample collection is within one organization. This is acting as a limitation to the study. If questionnaires are issued on a larger scale, the results might vary a little bit. Moreover, future research may be needed to find more reliable data and relationships as there is very little prior data and research available on this topic. Furthermore, the research was conducted using questionnaires that may not be implied later when new issues in the current study might emerge. Lastly, in this study, the concept of preferential treatment has been determined as a general concept of Artificial Intelligence. In future studies, the effects of these concepts might be examined separately.
2- Findings, Discussions, Conclusions, and Recommendations
The research was conducted through the distribution of questionnaires among the students at the institute. A total of 50 questionnaires were distributed among students. Their response has been broken down into a few parts below:
2.1.1- Awareness of the Technology
The questionnaire first asked the respondents if they were aware of the level of technology. According to answers, only 48% had neutral knowledge. The other 14% did not have much knowledge. Whereas 26% had little bit idea of technology and only 12% were proficient.
Figure 1: Percentage of People Aware of the Artificial Intelligence
2.1.2- Impact of AI on Business
According to collected data, 41.3% of people think AI will affect marketing.12.8% will affect matchmaking and registration process. And 23.4% of them voted for pre-show user behaviours. Only 10.4% think AI will affect sales.
2.1.3- Implementation of AI technologies
The questionnaire asked respondents if they are aware of any regulation that affects the implementation of AI technologies in business 36% of the respondents replied in the affirmative. And 38% replied no whereas and 26% voted for maybe.
The data collected through questionnaires gave several results. According to the results, more than half of the respondents were aware of the AI business. This means that AI has reached a stage where almost everyone is aware of it.
Moreover, the results show that AI is being practised the most in online business. Amazon has used an AI system in their fully automated stores. Whereas in our country AI is not used frequently in business as a businessman are not aware of its advantages. And how they can prevent loss in business using AI. Due to this, our country is lacking behind in all fields.
This might be true due to the limitation of the research within the institute only. The results might vary if the research is conducted on a larger scale. If the research was conducted on a larger scale, the percentage might have increased.
Regarding the use of AI in Pakistan business, the results state that almost half of the respondents think that AI should be used in business. However, this result might vary too based on the nature of respondents.
In the light of the results achieved from the research, several conclusions can be drawn. The technology of AI exists in all fields in Pakistan, but to a limited extent and everyone is aware of it. The Government is the backbone of a country in all its matters. The young generations have an interest in technology and are aware of Artificial Intelligence uses in every field. But they are unable to implement AI in most of the cases because of limited resources.
Following steps can be taken to root out the problems in Amazon business using AI technology slowly:
- Just putting brand logos with few features is not enough to compete with the physical stores that most people are used to shopping at so there is a need of some new features competitive enough to match the level of physical stores.
- Sometimes people need things urgently like someone needs an ingredient for a cake they will get goods in a few days. So, a quick way would be to use drones and other AI robotics for finding and delivering light weight things.
- The Prime Air project has been working for many years as Amazon needs to solve many problems like fear of drones falling from the sky etc. The possible benefit of Prime Air is driving more orders and even becoming an isolated service that Amazon can offer other businesses for the delivery of small and urgent items, which are: critical to Amazon’s future business.
- Amazon cannot hire enough people to process millions of orders. If Amazon wants to continue progressing it needs to find ways to scale and optimize its physical operation. Instead of asking human workers to walk miles each shift, Amazon must work on robots that drive the bins to store products.
•ML guy,5 Big Business Problems Amazon is Trying to Solve with Machine Learning, July 16 2018
•Akash Pandey, How Amazon gets benefits from Artificial Intelligence, Oct17 2020
•Firas Ghunaim, E-Commerce: AI and the Convenience Industry, Sept16 2018
•Clifford Dolan, Amazon’s Artificial Intelligence Application Is at The Forefront of a Commercial and Technological Revolution, Feb21 2018.
Ahmed Bin Tariq
Mahnoor Safeer Abbasi