EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

725. Forecasting B2B markets with machine learning and predator-prey models: empirical evidence from global Aerospace companies

Invited abstract in session TB-45: Artificial Intelligence and Machine Learning for Decision Support, stream Decision Support Systems.

Tuesday, 10:30-12:00
Room: 30 (building: 324)

Authors (first author is the speaker)

1. Dina Litsiou
Marketing, International Business and Tourism, Manchester Metropolitan University
2. Chaitanya Kapoor
University of Manchester
3. Konstantinos Nikolopoulos
Marketing and Management, Durham University
4. Chrysovalantis Vasilakis
Bangor University

Abstract

In this world of such a volatile, uncertain, and changing environment and markets, especially in a post-pandemic ‘new-normal’ context, it is very important to be ahead of competitors so as to maintain one’s market share. This is true for all markets, but more than anything for B2B markets that are affected immediately form supply chain disruptions and need to employ multiple channels to survive and thrive in turbulent times. Products and services can be sold profitably only if customer needs are satisfied at acceptable (low-ish) prices. Many factors are involved in making a product cost efficient, one of which is forecasting. Forecasting is one of the most essential parts of almost every supply chain in the world. This paper contributes towards the improvements in the forecasting methods currently being used by forecasters in B2B markets, a key area in the early days of Industrial Marketing Management (70s and 80s) that has been revived recently due to the revolution of big data predictive and prescriptive analytics. This paper is focused on forecasting the competing product life cycles using Lotka-Volterra predator-prey equations. The objective of this paper is to find a better version of Lotka-Volterra equations suited for general market products. Various regression analysis techniques have been applied in order to make an attempt to find the best way of forecasting using Lotka-Volterra equations.

Keywords

Status: accepted


Back to the list of papers