Listed below are a few of the recommendations for automated solutions working short SLA (Service Level Agreement) or real-time requests :
Since Blue Prism follows the practice of never finish the work instantly, and always follows a queue to process the work via the user interface, you can separate the work request confirmation from the work. For example, if a work request comes via a Web Service call, the Blue Prism must add the request to a Work Queue and give an output that the request has been received.
You can also allocate enough robots to ensure that those robots work with the work queues at peak times, with the expected SLA. Apart from this, a method of controllers must be designed into a solution to inform the users if an SLA is being missed or if its at risk. Work queues can be also reported up to estimate whether the available capacity is enough to meet the expected workloads at peak time.
Robust contingency can be met, as it is an integral part of the Blue Prism framework. Here, you have to make sure that the expected SLA are included in the contingency plans.
