There are countless reports, articles and mentions about how the use of analytics based on Artificial Intelligence (AI) and Machine Learning (ML) will be able to help in the demand planning processes, improving the accuracy affected by the increase in complexity in the portfolio, in the distribution channels and, also, in the competitive environment.
Today, in most companies, the size of the team responsible for processing and analyzing data with traditional time series tools and regression models imposes a limitation on the possibilities of analysis, forcing the grouping of series for forecasts. top-down, restricting the number of exogenous variables inserted in the model and the treatment and maintenance of the baseline of few series, which impairs accuracy.
But what can we expect from these new solutions?
Firstly, it is necessary to understand that the improvement in accuracy will come from aspects of Increase e Automation of the demand planning process. Increase refers to the growth in the volume of processed data and variables considered in the models, allowing the identification of imperceptible patterns to the “human eye”. Already Automation discusses the possibility of automating the processing of sales series data, correcting the baseline and using more granular sales information, which without proper adjustments results in greater errors.
In addition to the improvement in the accuracy of the monthly tactical plan, we can expect the identification of long-term consumption trends, helping to define the portfolio and the development of new products, as well as the sophistication of very short-term replenishment mechanisms, with demand sensing.
However, there are still significant barriers to the adoption of AI and ML in the demand planning process, such as the low maturity of the current process, lack of technical knowledge and integration in the chain to obtain data, among other aspects, which I intend to address in a next post.
And your company? Already using AI and ML engines in the demand planning process? Share with us on the ILOS social networks! Big hug!
Leonardo Julianelli
https://ilos.com.brExecutive Partner of ILOS. Graduated in Production Engineering from EE/UFRJ, Master in Business Administration from COPPEAD/UFRJ with extension at EM Lyon, France, and PhD in Production Engineering from COPPE/UFRJ. He has several articles published in periodicals and specialized magazines, being one of the authors of the book: “Sales Forecast: Organizational Processes & Qualitative and Quantitative Methods”. His research areas are: Demand Planning, Customer Service in the Logistics Process and Operations Planning. He worked for 8 years at CEL-COPPEAD / UFRJ, helping to organize the Logistics Teaching area. In consultancy, he carried out several projects in the logistics area, such as Diagnosis and Master Plan, Sales Forecast, Inventory Management, Demand Planning and Training Plan in companies such as Abbott, Braskem, Nitriflex, Petrobras, Promon IP, Vale, Natura, Jequití, among others. As a professor, he taught classes at companies such as Coca-Cola, Souza Cruz, ThyssenKrupp, Votorantim, Carrefour, Petrobras, Vale, Via Varejo, Furukawa, Monsanto, Natura, Ambev, BR Distribuidora, ABM, International Paper, Pepsico, Boehringer, Metrô Rio , Novelis, Sony, GVT, SBF, Silimed, Bettanin, Caramuru, CSN, Libra, Schlumberger, Schneider, FCA, Boticário, Usiminas, Bayer, ESG, Kimberly Clark and Transpetro, among others.
