3 Laws of AIM
Inspired by Isaac Asimov's 3 laws of robotics I have engineered my own 3 guiding laws to ensure Asset Information Models (AIM) are assembled coherently for this age “the internet of BIM” where data is at the centre.
AIM is orientated around the built environment and an AIM may contain all the data for physical assets within a businesses buildings and infrastructure. All of a businesses physical assets serve the business to enable an environment for it’s activities ranging from services to production.
In this era of Data documents, 3d and 2d models, reports are views of the data and applications are conduits to the creation and manipulation of the data.
The era of “product” saw barriers to interoperability of data due to proprietary data formats for each application. In the era of data there is no proprietary format for the transmission and sharing of data. All data storage and retrieval systems must accept and transmit data in open formats. Data transmissions can be encrypted and unencrypted for end to end security but the formats of transmission are still the same open formats.
Law 1: Relevance
All data must be relevant to achieve a declared outcome. Production of data that has no purpose has no value.
Law 2: Accuracy
All data must be accurate and this has to be verified. If the data does not have an author then it cannot be verified and therefore trusted, it conflicts with the first law of Relevance. When it can be verified if has to pass verification process to be trusted. This can be achieved by peer review and machine learning algorithms. Machines can easily test issues like existence, data type, unit type and within permitted values.
Law 3: Accessibility
All data must be gathered at speed so that it can be used in part, parts or whole in open standard’s this means it has to accessible. By adhering to the 1st Law of Relevance less data is easier to gather than more data. Gathering data at speed must comply with the 2nd law of Accuracy and the speed of verification is critical to making this practical. Data only needs to be verified once. If the data is copied then checking if has not been corrupted by accident or by malicious intent and accessible to a verification service is fundamental.
If these 3 laws are followed then a whole new world of automation opportunities opens up.