Result
Partnering with Plavno, the customer received an Al-powered Demand Forcasting solution that reduced the cost of ownership by 50% compared to the previous solution, reduced the time to process information and get a forecast by three times, and increased forecast accuracy by 20%.
Customer
LPlast company dominates the market of vinyl siding manufacturers in Poland. The company produces a huge range of products, numbering more than 1500 SKU. In addition to the production of vinyl siding, the company operates its own retail stores network in all regions of Poland. Their total number is 143. The number of employees exceeds 5000
Challenges & solutions
Our client's market is the construction market. LPlast is one of the first tiers in complex global supply chains dealing with large volumes of goods. This situation leads to unstable demand patterns that are particularly difficult to interpret for forecasting models. The client's ERP system allows saving sales history only for up to 3 years. Unfortunately, this 36-month limitation is all too common in supply chains and is a significant obstacle to achieving the highest possible accuracy
To ensure we maximized our chances of success, we tried models based on both machine learning and statistics. The machine learning model ran on historical data that was sourced from the client's ERP system. Our statistical models rely on three advanced concepts: multiple temporal aggregation, improved seasonality detection, and pattern assembly
LPlast is currently reviewing its demand forecasting methods. The company’s existing forecasting system was frequently incorrect in its forecasts. The poor quality of forecasts led to decrease in the company’s profits, constant overstocking and understocking. In addition, the existing demand forecasting solution was slow and expensive for the company
As a result of our team’s work, we were able to obtain more accurate data for a period of 6 months compared to the system the client was using. In addition, our solution was able to provide 10,000 forecasts in less than a minute. The previous solution could deliver up to 3,000 forecasts per minute. In order to address the high cost of service challenge, we utilized cloud technologies
proccess
Agile methodology forms the cornerstone of our work philosophy. Through a seamless blend of innovative practices, including regular demos, comprehensive progress tracking, and a unique pay structure based on hours invested, we've constructed a workflow that ensures optimal outcomes and client satisfaction.
project team
Looks like your business could benefit from a similar service? Let’s discuss how we can help you reach your business goals!
Plavno experts contact you within 24h
Submit a comprehensive project proposal with estimates, timelines, team composition, etc
Discuss your project details
We can sign NDA for complete secrecy
Vitaly Kovalev
Sales Manager