TY - JOUR
T1 - Prediction-led prescription: Optimal decision-making in times of turbulence and business performance improvement
AU - Schäfers, Andreas
AU - Bougioukos, Vasileios
AU - Karamatzanis, Georgios
AU - Nikolopoulos, Konstantinos
N1 - Data availability statement: Data will be made available on request.
PY - 2024/9
Y1 - 2024/9
N2 - Can you have prescription without prediction? Most scholars and practitioners would argue that a good forecast drives an optimal decision, thus promoting the concept of prediction-led prescription. In times of turbulence, Special events like promotions and supply chain disruptions are impacting businesses severely. Nevertheless, limited research has been carried out to date to accurately forecast the impact of, and consequentially prescribe in the presence of special events. Nowadays Artificial Intelligence (AI) predictive analytics methods and heuristics imitate and even improve human intelligence, progressively leading towards innovative cognitive analytics solutions. This research aims to contribute to applying advancements in AI-based predictive analytics to improve business performance. We provide empirical evidence that these AI solutions outperform the popular (especially among practitioners) linear regression models. We corroborate the stream of literature arguing that AI predictive analytics could − via a natural path-dependent process − enhance prescriptive analytics solutions, and thus improve business performance.
AB - Can you have prescription without prediction? Most scholars and practitioners would argue that a good forecast drives an optimal decision, thus promoting the concept of prediction-led prescription. In times of turbulence, Special events like promotions and supply chain disruptions are impacting businesses severely. Nevertheless, limited research has been carried out to date to accurately forecast the impact of, and consequentially prescribe in the presence of special events. Nowadays Artificial Intelligence (AI) predictive analytics methods and heuristics imitate and even improve human intelligence, progressively leading towards innovative cognitive analytics solutions. This research aims to contribute to applying advancements in AI-based predictive analytics to improve business performance. We provide empirical evidence that these AI solutions outperform the popular (especially among practitioners) linear regression models. We corroborate the stream of literature arguing that AI predictive analytics could − via a natural path-dependent process − enhance prescriptive analytics solutions, and thus improve business performance.
KW - AI
KW - Forecasting
KW - Prediction-led Prescription
KW - Special events
KW - Turbulence
UR - http://www.scopus.com/inward/record.url?scp=85197301508&partnerID=8YFLogxK
U2 - 10.1016/j.jbusres.2024.114805
DO - 10.1016/j.jbusres.2024.114805
M3 - Article
VL - 182
JO - Journal of Business Research
JF - Journal of Business Research
SN - 0148-2963
M1 - 114805
ER -