Prescriptive Analytics

When knowing what will probably happen, a next logical step is to define a set of actions and analyse the probability that an action will result in the required effect. The predictive analysis is extended by comparing the consequences of actions based on a predictive result.

If the best action is automatically invoked, for example by using an event processing environment, your company has reached a level of dynamic adaptation without human interference.

Your organisation is really becoming agile when this automated prescriptive analytics level is reached.

Monitoring of the actions is needed to determine whether the models in use are still valid or should be fine-tuned or altered.

Our data scientists will help you design the statistical models as well as will monitor and improve these models during use.

Your organisation is really becoming agile when this automated prescriptive analytics level is reached.

Data Sources

To be able to predict, the scope of data sources should be a subject of investigation. Do we have all the required data? Could we use data from external sources and in real time? Using moving data in streaming analytics will open up all kinds of new solutions.

Prescriptive analytics will almost always trigger an automatic follow-up. Based on the probability of the predicted outcome of actions the best action is automatically executed.

Models and Monitoring

Predictive models are defined in a joint operation between business analysts and data scientists. You need to understand the business to be able to collect the right data and define models using this data. Prescriptive models will give feedback on pre-defined actions and calculate the probability that a specific effect will be realised. This will be the input for the decision-makers.

Our analytics architects will organise a monitoring loop in order to fine-tune and adapt the prescriptive models.

From Predictive to Prescriptive Analytics

The combination of predictive and prescriptive analytics will help you achieve both efficiency and effectiveness. For example, predictive analytics will help you understand the drivers behind customer buying patterns to anticipate the products customers want. Prescriptive analytics will help you optimise scheduling, production, inventory and supply chain design to deliver what they want in the best way.

From Prescriptive to Streaming Analytics

With prescriptive analytics next actions are evaluated and the best action will probably be executed. It makes no sense to execute automatically the best action if there is no urgency. Urgency originates when moving data needs to be analysed in real time and used in the decision-making. Moving data in streaming analytics will open up all kinds of new solutions. Matching customers’ behavioural history to a store’s inventory will help promote specific products.

Know your customer and act within seconds!

devoteam

Contact

Darrell Woods

Senior Account Manager – Enterprise, Devoteam UK