What is a recommendation?

Smartlink’s recommendation-feature is an advanced way of equipment monitoring. Instead of relying only on diagnostic experts, we now also use big data science and algorithms to spot (critical) points of improvement. 

Based on algorithms and daily calculations of our data science team, Smartlink generates recommendations. Such suggested interventions can prevent shutdowns, improve energy efficiency, or prevent huge repair costs among other things. 

It is an important step in the realm of predictive maintenance.

How it works

Your equipment sends data to our connectivity platform regularly. This could be sensory data, counter values, warnings, shutdowns, and machine settings. In addition, we can also use administrative data (service history, bill of materials, and data from Service Connect, etc.).

All this data goes to the cloud. Once our big data team processed this raw data, algorithms are applied. This results in the recommendations you receive in Smartlink.

Where to find recommendations

Smartlink shows active recommendations in different places. The first place where you can find it is in your customers' dashboard in the event widget.

This widget is also being used to collect feedback. Customers can like/dislike recommendations and provide additional feedback. We use this information to improve the accuracy of the recommendations we show.

You can also find the recommendations in the Events-overview on the left menu.

NOTE: the message of recommendations is usually not the same for external users. Internal users see the same recommendation but with more detail and context. However, when you use the site dashboard, you will see exactly the same recommendations as your customer.

The recommendation overview (admins)

The recommendation overview gives you complete oversight of the smart events that were triggered in your customer center, including the historic recommendations that were published. It is similar to the overview available to customers. However, it contains extra information (such as the historic recommendations).

In this overview, you can see which recommendations are visible to customers or distributors.

Not every recommendation is visible to all user roles. It mostly depends on the type of recommendation and the type of user (internal/external). Based on the recommendation’s relevancy to the user, we show it to a selection of user-roles. The relevancy largely depends on whether the recommendation concerns diagnostic (DiaEq), internal (IntEq) or external equipment (ExtEq). These recommendations are then visible to diagnostic center users, internal users, and external users respectively. You can filter the overview to these recommendation-types.

NOTE: the principle of license and contract dependency also applies to this feature. In other words, what external users can see depends on their active contracts and licenses.

The recommendation remains active for a predefined period allowing the user to see it. When the predefined period has expired and the recommendation is no longer active, the recommendation disappears. You can check the expiration date in this column:

NOTE: Right before the recommendation expires, we check the conditions that triggered the event. If the conditions have not changed and the recommendation still applies, the expiration time will be extended.

How to disable a recommendation

You might prefer to hide certain recommendations from your customers. If you want to do this, go to the recommendation overview for admins, select the recommendation(s) and use the action menu to disable the recommendation.