Ailys develops a Demand Forecast Model with Pulmuone Foods

Lewi Kim
ailys
Published in
3 min readJan 24, 2022

-Aims to reduce financial loss in domestic food distribution market and improve decision-making issues concerning food demands

-It is notably exceptional to have a domestic demand forecast model run with machine learning

-Savings of billions are expected throughout the industry by the establishment of a production system based on sales

Ailys has signed a contract with Pulmuone Foods through the 2021 AI Voucher Support Project introducing DAVinCI LABS, an optimizing machine learning solution.

This year’s AI voucher support project supports the use of AI solutions and services for small and medium-sized enterprises and ventures seeking introduction of AI technology by the Ministry of Science and ICT(MSIT) and Korea Electronics Association (KEA). We decided to develop an AI-based prediction model that forecasts demands of food items.

Trials to build a prediction model based on AI have been limited to financial domains, not to mention the apparent lack of machine learning use cases within the domestic distribution industry. If the number of categories or cooperative companies exceeded a certain point, hurdles in data preparation and management that were required as a process of machine learning analysis arose. There also existed a problem that without a data expert, developing an advanced model that reduces the error gaps of demand forecast was considered extremely hard to succeed.

However, Ailys is planning to expand the range of prediction management to all categories with the developed demand forecast model through this project. Meanwhile, Ailys strives to utilize DAVinCI LABS’ automation function to reflect on exterior variables and data used for special events like promotion discounts. The objective would be to strengthen and heighten the accuracy of the model.

DAVinCI LABS’ time series analysis helps create a demand forecast model.

Situations in the food industry are extremely turbulent from day to day, which means there’s an apparent need to reflect recent data constantly. Along with this the following should be considered: time series analysis algorithms that detect change according to time, and supervised learning algorithms that take into account all variables that have an impact on the demand. Such advanced technique was barely available nor approachable with general coding skills, but with an automation solution it turns into a palpable concept.

Ailys, a technology company specializing in machine learning, developed an artificial intelligence solution ‘DAVinCI LABS’ and curated real business cases and examples over the finance industry, ranging from #insurance underwriting, credit scoring system (CSS), credit card issuance review, #fraud detection service (FDS), and customer relationship management (CRM) even to marketing. Overseas, larger corporations such as Mitsubishi Corporation and Aeon Financial Group have adopted DAVinCI LABS as a standard platform for data analysis. In addition, it is being used in various tasks of vast industry fields, such as demand management, target marketing strategy, credit scoring, fraud detection in manufacturing, distribution, and logistics.

Ailys already holds a previous record of utilizing DAVinCI LABS to develop a prediction model for demand forecast and stock management. That’s precisely why we could take part in this project with confidence. We will continue to apply DAVinCI LABS onto various domains of domestic distribution business and draw a notable business output of work efficiency enhancement and cost reduction.

--

--