Shionogi built AI-SAS on SAS Viya to semi-automate clinical trial programming and cut about 100 hours per trial.
Routine analysis programs were rebuilt for every trial, costing hundreds of hours each. Shionogi built AI-SAS on SAS Viya, using machine and deep learning to semi-automatically generate SAS programs. The page reports a 30% cut in analysis workload, roughly 100 fewer hours per trial (350 to about 250), and adoption by other Japanese pharma firms and CROs.
SAS is an analytics and AI software vendor, used across pharma, hospitals and medtech for clinical trial programming, prediction and decision support
This sits in the Ship stage of the AI use-case lifecycle: Named examples from the vendors pharma teams use to get an AI system live inside a real, regulated production workflow — validation, CRM, safety and clinical systems of record.
Published by SAS on its own site — read the original case study on sas.com ↗
Related use cases
- 01Dompé industrialised AI model deployment for drug discovery on SAS Viya.
- 02A New Jersey children's hospital cut no-shows 8.5% with a SAS Viya-based predictor.
- 03Erasmus MC set up a responsible-AI lab with SAS and Delft to support ICU ventilation decisions.
- 04Frankfurt University Hospital uses SAS Viya on AWS for antibiotic stewardship and resource forecasting.
- 05Siemens Healthineers predicts lab analyser failures on its Atellica solution, improving uptime by 36%.
- 06Transforming Content Operations for 2x Faster Time to Market