How to Consider Use Cases in Life Sciences for Machine Learning
Empowering business growth with disruptive technology like IoT, predictive analytics, and artificial intelligence has become a norm in IT, and machine learning is leading the way as software applications are becoming smarter to improve our corporate and personal life. With massive improvements in hardware for big data use cases; machines can sense, understand, interact, predict and respond to solve an industry business problem. Bio Pharmaceutical brands are very critical intellectual property and optimizing your marketing ROI through brand intelligence and sponsorship insights is a powerful use case that can improve brand recognition. Similarly, service ticket intelligence is very important in medical devices because automating the classification of errors, issues and suggested responses of customer support tickets can help to improve service levels.
A few key questions can help determine whether a use case is fit for machine learning. For example, are you able to automate the high-volume task? Is there a pattern involved in the business process having unstructured data sets? Enterprise data is transformed into business value with the help of a model, by using input and output parameters. Predictive models may have some bias with respect to the degree of a model fitting the data and certain amount of variance based on how a model can change with parameters. Below are few use cases by line of business:
- Quality needs to be enforced in Supply Chain and manufacturing business process for regulatory compliance aspects and root cause analysis is a key aspect of Corrective and Preventive Action (CAPA), which aligns with industry initiatives like QbD (Quality by Design), PAT (Process Analytical Technique), and CPV (Continued Process Verification). There is a clear need to identify main causes for the reported defects under a material asset category and understand the percentage impact of identified causes to manage the overall defect count. Based on gathered data, machines can predict what can be produced vs planned production for a specific duration based on historical production information, thereby preventing these deviations and non-conformances. Analyzing the cause of deviation of cycle time from standard cycle time for manufacturing equipment and prescribing measures to achieve cycle time that is closer to standard cycle time has an impact on yield and scrap.
- Life Science companies have a huge spend with direct and indirect materials and services with contract organizations. Machine Learning services are relevant for commodity managers who focus on the optimization of global spend. There are various machine learning use cases in strategic sourcing and procurement, such as assessment of contract negotiation behavior, optimization of awarding contract to suitable candidates, detection of single sourcing risks and determination of how much component sourcing to outsource to contract manufacturers. Intelligent Enterprise can recommend possible suppliers whenever customers may need to replace an existing supplier who is not performing as expected, replace a supplier that poses a compliance risk, select additional suppliers to comply with purchasing policies, expand to a new territory or category of spend, or find cheaper options for materials/services.
- Learning management is critical in regulated industries as training is a big part of human resources and act as shared services in Life Sciences. HR business partners can narrow down the shortlist to find the best candidates by parsing resumes into structured information, visualize candidate profiles by skills, education, and experience, so that he/she can compare and generate best fit score of profiles to jobs and vice versa. Talent management can take a more personalized approach towards career mapping based on employee’s unique situations to make well informed better career decisions based on skill trajectory, and learning sessions taken by an employee thereby opening opportunities for fast track growth path.
- Consider use cases where matching algorithms are used extensively for shared services like cash Matching incoming payments with invoices is now a simplified process adopted by Intelligent Enterprises as an asset for clearing volumes of backlog data. Machines can match accounts receivable invoices based on learned criteria i.e. action of accountants and provide a confidence score thereby helping accountants to clear payments faster if the matching rate is within threshold. For payments that cannot be cleared automatically due to lower confidence levels, accountants can see a list of the best fitting invoices, saving time to identify the relevant receivables.
- Similarly, an accounts payables accountant needs to release payment blocks for supplier invoices, to get the invoice paid and receive cash discounts when paid before deadlines. Based on historical data, current user interaction and machine learning algorithms, the system can react automatically or suggest resolution proposals to the user. Decision basis may include supplier rating, deviation vs. cash discount achievable, or purchasing category. Matching invoice line items with purchase order line items, and providing remittance advice to reduce manual errors are additional use cases relevant in the Life Science industry, as automation enables scale in shared services model.
- Sales and marketing has a big potential for Life Sciences manufacturers to leverage machine learning during sales negotiations with wholesalers, hospitals, clinics, and retail pharmacies by capturing keywords, sentiments, competitors, and new contacts to feed into deal scoring and ultimately improving the win rate. Bio-Pharma Sales Reps can share relevant marketing collateral with Physicians and key opinion leaders based on their interest level. Segmentation is key in Marketing to create target groups for campaigns based on the 3rd party prescription data that provides behavior data to boost sales. Thus, machine learning can help build customer loyalty with proactive retention strategies in the Life Sciences industry.
In summary, SAP offers machine learning services as part of our Leonardo IoT platform for faster decision making i.e. automate and prioritize routine decision making process to achieve best outcome sooner and discover and process new insights to adapt to rapidly changing business environments. For more information on SAP machine learning, refer our white paper ( https://www.sap.com/documents/2017/05/de7cfb6d-b97c-0010-82c7-eda71af511fa.html ). Smart business process enabled by machine learning from SAP can help to achieve the goal of Intelligent Enterprise for the Life Science industry. In the coming months, selection criteria will evolve as the IoT technology adoption rate improves. Artificial intelligence and Enterprise Bots have potential to boost the automation in business process to the next level sooner than later and we hope to see more reference case studies in Life Sciences.