-
Lentia
PHASE I PROPOSAL
Name of the proposed solution
Lentia
Problem statement
Nowadays, businesses lose much money globally due to excess or deficit in inventories. Despite efforts, traditional methods of demand forecasting are not efficient enough.
Motivation to choose this problem
While the current global debate about how the world will look like after COVID-19 highlights the great concern about the future of economy and employment, it is now crucial to reimagine the postpandemic business environments to prevent any sudden shutdowns and support the running businesses as much as possible to help them reduce costs and make profits.
Solution statement
AI demand forecasting tool that helps to reduce errors in traditional demand forecasting methods.
Other solutions
A platform to connect buyers and suppliers by asking all potential buyers about their future purchase plans and then give this data to suppliers and retailers compared to previous month data to put some tolerance in production and satisfy customer needs. On the other hand, customers who fill out this required data on the app will be rewarded by suppliers.
Solution Pitch
As a solution to the previously mentioned problem, demand forecasting will be the keyword. Since demand forecasting is a field of predictive analytics, AI and machine learning could be used. Using machine learning especially deep learning will help to reduce errors by 30 to 50% in supply chain networks. In the data science and AI field, there are two programming languages used mainly, R and Python. In our solution, we will rely on Python programming language as the core and this will lead to the next step. Specifically, we will use deep learning to handle and analyze this huge amount of data and get accurate outputs. To achieve this, we will use some other tools besides Python itself: NumPy: A library Python, adding support for large, multi-dimensional arrays and matrices, along with a large collection of high-level mathematical functions to operate on these arrays. TensorFlow: An end-to-end open-source machine learning platform. Cloud AutoML: A suite of machine learning products that let training high-quality models. It relies on Google’s state-of-the-art transfer learning and neural architecture search technology. Keras: A high-level neural networks API, written in Python and capable of running on top of TensorFlow, CNTK, or Theano. Moving now to the client-side, we will allow our clients to monitor and analyze data in a user-friendly way as well as get a full control panel to control almost everything. Our Solution will help improving manpower planning which will optimize costs and HR effectiveness. It will also have an impact on customers not only retailers and businesses, demand forecasting in general increases customer satisfaction, and this will of course improve brand perception and increasing customer loyalty. According to Mckinsey Digital, the estimated impact of AI in the supply chain is between $1.2T and $2T in manufacturing and supply chain planning.
Most valuable customers (MVCs)
As the issue of supply chain deficiency mainly faces businesses with huge production capacities and since the pandemic greatly affected the demand on their products negatively, it is important to solve their supply management problem to cut the budget waste by providing them with reliable information to make decisions accordingly. Improving the accuracy of demand forecasting by reducing errors of traditional statistical methods will greatly contribute to cost reduction, optimal inventory planning and in return help these businesses survive during the demand shocks caused by the pandemic. To further understand the needs of big companies as our target customers, surveys, in-depth interviews with managers and retailers, and feedback forms will be provided to make sure that our approach matches their needs. Addressing the need to increase the reliability of demand forecasting outcomes will be achieved through the use of deep learning to reduce errors by 30-50%. Throughout our project, we focused on creating incentives to attract big companies such as the reliability of the software, the full control panel provided, and saving manual computations time.
Innovation and value proposition
We provide an accurate demand forecasting tool with full control panel and time series analysis. Moreover, we make affordable plans for small businesses to help them improve their customers' satisfaction and generate more profit. Along with our AI demand forecasting tool for business, we provide consumers with info about the supply of certain products, especially at times of crisis, since they cause high fluctuations in demand, and create limitations in production. Furthermore, we offer them plans to meet their needs for certain products, without causing a shortage of these products in their region.
The core technology powering the solution
In our solution, we will rely on Python programming language as the core and this will lead to the next step. Specifically, we will use deep learning to handle and analyze this huge amount of data and get accurate outputs. To achieve this, we will use some other tools besides Python itself: NumPy: A library Python, adding support for large, multi-dimensional arrays and matrices, along with a large collection of high-level mathematical functions to operate on these arrays. TensorFlow: An end-to-end open-source machine learning platform. Cloud AutoML: A suite of machine learning products that let training high-quality models. It relies on Google’s state-of-the-art transfer learning and neural architecture search technology. Keras: A high-level neural networks API, written in Python and capable of running on top of TensorFlow, CNTK, or Theano.
The impact of your solution on people’s life
Our Solution will help improving manpower planning which will optimize costs and HR effectiveness. It will also have an impact on customers not only retailers and businesses, demand forecasting in general increases customer satisfaction, and this will of course improve brand perception and increasing customer loyalty. The estimated impact of AI in the supply chain is between $1.2T and $2T in manufacturing and supply chain planning.
Business model
Customer Segment: B2B including: Companies, Start-ups, and Other Businesses
Channels: Website, and Headquarter
Revenue: Subscriptions
Benefits: They will reach to customer satisfaction and reduce the percentage of waste and surplus in inventory that doesn’t use. Reduce cost that results from excess in inventory Especially during a period of sudden changes such as "Virus Corona", which affected production, as it led to increased demand for products and less demand for others, which negatively affects factories, companies, and organizations, which leads to closures in some cases.
Sustainability
Since the problem has existed even before the COVID-19 pandemic, then it is a sustainable problem to work on. Our solution can be applied even without any special circumstances as it helps businesses to scaleup as well as prevent huge numbers of money loss. We will continue offering our services normally after this pandemic is over.
Scalability
In the future, our solution has a high scalability range because of the field we are working on. The supply chains are going to be bigger; the demand is going to increase, and Machine learning has a lot of research in it which evaluates constantly. By, all these changes in this field, we always need to change and to accurate our results. One of the strategies we want to apply to our project is to apply certain encryption scheme which allows any company to use our service without reviling it data maintaining the quality and the accuracy of the service.
Log in or sign up for Devpost to join the conversation.