Publications and deliverables

Deliverable 5.2

Title:

Execution Engine Prototype

Due date:

2021-02-28

Executive summary:

The goal of this deliverable is to select a suitable rule execution engine for integration into the HumBAS landscape. Another focus is the implementation of this rule execution engine into the existing software prototype, namely the HumBAS-server. Based on the results obtained from the previous deliverable (see Deliverable 5.1), identified rules are transformed into a machine-readable format. Most basically, a rule execution engine consists of a production memory containing the rules and a working memory, which compromises of facts (data). These two will be combined in a pattern matcher of the rule engine and trigger events based on the result. Several engines were considered, however, as Drools provides a Spring Boot integration into the existing Java prototype and human-readable rule files, it was chosen as HumBAS rule execution engine. The rule execution engine has to consider several aspects before and after evaluating feedback from users. Foremost, the most current sensor values are integrated into the feedback evaluation. Depending on these values and configuration parameters, the rule engine triggers an action for the \gls{bas}, requests feedback from other feedback tools, informs the facility manager or does nothing. The rule execution engine does only request feedback from other feedback tools, if the feedback initiator is in an open-plan offices where multiple feedback tools are online. Then any positive action (e.g. changing the shutter position, adapting the temperature) will only trigger, if a certain amount of users do response similar to the feedback initiator. Either way, the user which gave the initial feedback will receive a positive or negative feedback response. This enables HumBAS to cover most cases of feedback a user can give and supports the facility management in processing this feedback.