Smart Data Validation – 2
Forms in Smart RDM – check how it will streamline your production.
I encourage you to read part one first, which is available at this link.
Data under control
Therefore, we already have an answer to the question of manual data entry and an explanation of how forms support data validation and regulatory reporting. However, these drop-down lists are not the only way to prevent incorrect data entry. To control correct data input, you can set up rules on the form to block the possibility of entering values of a different type or outside a specified range. For example, information on the destruction of a piece of equipment must be completed on a scale of 0 to 10. If the operator enters “100”, an “out of limit” message will appear.
One method of validation is also to define mandatory fields: this makes it impossible to save the form if the required fields (parameters) are not filled in. For example, a form consists of 10 parameters, 3 of which are key. If these are left ‘blank’, the form will not save, even if all the non-obligatory fields are completed.
Different validation methods can therefore be ‘sewn’ into the forms, and access to them can be managed – which is also a guarantee of data correctness. What does this option consist of? The form consists of a set of specific parameters, and which parameters are filled in and by whom depends on the user’s rights, as selected parameter groups can be assigned to particular user groups. How does this work in practice? Let us assume that a process consists of laboratory and production data. Users working in the laboratory can edit only laboratory data and production employees – only production data. At the same time, data for the entire process can be visible to all process participants – sometimes, it is impossible to complete information without checking the broader context.
More advanced access settings are also possible. For example, the machine operator will not be permitted to edit process data (i.e. that which comes directly from the control/automation systems), even if it is erroneous. Instead, they can enter the comment “dirty sensor”, which is non-process data (i.e. describing a process, job number, tonnage to be produced, and comments). As you can see, changing the process data would make it impossible to fix the cause of the problem in production.
A tool that also enables data verification is the acceptance process, or so-called workflow, sewn into the Smart RDM. The standard approach consists of the following stages: edit, accept, approve/reject. At each stage, a history of changes is recorded, along with information about the user who entered the data. Once the acceptance process is complete, the document can be published, e.g. in pdf format and distributed to subscribing users. Accepted forms are also the basis for the automatic generation of reports, ensuring that they are prepared in accordance with the applicable standard. A good example is environmental reports, which must be prepared in accordance with guidelines. Another is TSO and DSO reports, which have many requirements that must be met. Acceptance process guarantees that all the reports will be created due to right standards.
Flexible and simple configuration and modification of the forms and their behavior
How are specific forms created? Their templates are just normal AF templates and can be created as a part of initial deployment of Smart RDM, when special Smart RDM AF model is setup, or added anytime later. No other skills than PI Asset Framework knowledge are needed. Set of parameters for the form is defined in AF and assigned to a resource, e.g. a machine or a process – all depends on the business requirements and is as flexible as PI System is.
The forms are generated automatically (or manually if required) within a specified timeframe. For example, if a form is assigned to a machine, which requires hourly resolution in reporting, Smart RDM can generate 24 forms (1 per hour) daily. The frequency/resolution of forms is defined in the configuration, e.g. hourly, per shift, daily, monthly, etc. Based on this, plans and schedules can be created. For example: with an administrator-defined resolution, we can enter a production forecast, based on which we can monitor the realisation of the production plan and adjust it continuously in the following months.
As an additional convenience, notifications can be assigned to forms. These feature allows to handle dynamically-generated forms, for example when some event occurred. In such case the notification may include a link to a form “waiting” to be completed – to help the user to enter the information in the right place quickly.
Based on the available data types, parameters can be defined in the form to create and update rankings. These can relate to various production indicators, for example, ranking of machine breakdowns, causes of breakdowns, working time, downtime, etc. The choice of indicators to be ranked depends on the customer’s need and the technical possibilities of obtaining the necessary data.
Say hello to forms, say goodbye to Excel.
To summarise: the forms in Smart RDM are a very powerful functionality, and manual data entry implies many possibilities and offers just as many benefits. Each piece of information in a specific form in combination with the data traditionally acquired by PI System allows to monitor, analyse and forecast the process.
Forms make the data complete, reliable, up-to-date and ready for reporitng (or exporting eg. to data lakes) – because they can be completed and verified on an ongoing basis in an orderly and controlled manner. At the end this allows to have a consistent and standardized process reports with no ad-hoc manual changes, lack of transparency, difficult data tracebility and scattered Excel files.
Probably every person reading this text encountered data chaos when “everyone wrote it down as they wished” – literally (sometimes even by hand in a notebook). Forms eliminate that problem as users get one tool and one place where data is collected and shared according to permissions and accepted standards.