Mr. Alex Binder ICONICS Senior Applications Consultant demonstrates how AnalytiX-BI and AssetWorX come together for a real application for a marine port example.

Video Transcript

[0:00] Mark Hepburn ICONICS Vice President of Global Sales

I would like to invite our colleague from Melbourne, Australia, Mr. Alex Binder, to share with us a project that he's built to demonstrate how all these things come together and bring value in a real application set. So, this is a marine port example. So, Alex, he has 14 years of industry experience; he's been with ICONICSS for a number of years. He is an expert in HTML5, and he’s built his own products. Great stuff. Alex, if you can join us, please. We have Alex on connection.

[0:43] Alex Binder ICONICS Senior Applications Consultant

Thank you for the introduction, Mark. I'd like to share with you how ICONICS AnalytiX-BI and AssetWorX both come together in a real application. ICONICS software is used on a number of marine port applications worldwide. In this case, I was asked to put together a conceptual framework for a port in Southeast Asia. Here we have a homepage display which is showing data on the top 50 busiest container ports and the container traffic over the last decade. I'm grabbing this data from a web service which is imported into the AnalytiX-BI module using the data flow tool. I can click through the dataset and see the growth of container traffic over the last 10 years. I can also filter by regions and ports in a specific region. And just by clicking on these buttons here, you'll notice that displays are very responsive. And this is because with AnalytiX-BI, the data is cached within a memory database. Another thing that's useful to note is the display uses only a single grid works chart which is linked to the single AnalytiX-BI data model. Now, the buttons on this display simply set global aliases to the analytics query used in the chart. So, this makes configuration of the display in GraphWorX very flexible while at the same time also remaining very easy to implement and maintain. Now if we drill down into one of the ports, I have an aerial view showing the layout of some different terminals here. In this demo, we have container terminals. We have a storage tank farm terminal, and we have refrigerated container or cold storage terminal. If you click on the transparent overlay, you can go into a particular terminal. And we see the map view of the equipment laid out over a map showing real time data. And we can also view some of the KPIs for that equipment. So, in this case, cranes overlaid on that cranes position, you can also access the different terminals through the tab view at the top. And if you operated multiple ports, you can add them as additional tabs here. Now let's go into the cold containers example. One of the great features in AnalytiX-BI which I find myself using again and again is the ability to do SQL ad hoc queries in the GraphWorX display. So AnalytiX-BI lets you query the data set. The time that container has been in storage, I can search for the minimum, its arrival time, departure time. I can do some of the metrics like the consumption per hour and storage. I can query the temperature data set, and I can say what is the maximum. And that's how it's linked to a single query on this tile here. It really simplifies my job of creating analytical displays. I can point everything on a display to the same data source, the BI table, and then just write different queries to it. And because the memory caching, the AnalytiX-BI’s responsiveness is great. Let's go through the cold containers display. Here we have the cold containers overview. It's got a list of refrigerated containers that have arrived at the terminal. And you can select them from a selection here. And it will show the summary for that container. And also, it's real time data like the per hour the temperature, humidity, energy consumption that refrigerated containers are using. We pull in the data such as when it arrived at the terminal, when it departed, the contents, the customer it belongs to, the container ID, all of that will come from any ERP system such as SAP, and we match this up against the PLC level data, which would have been logged historically in the ICONICS Hyper Historian. And it's all pulled together into an AnalytiX-BI data model using the BI data flow tool. This tool is also used to clean up the datasets; you might need to change data types to match things up, to transpose data. You can tidy up some of the string formatting. And it allows us to create displays that gain insight into all of this. So, for example, here I have a refrigerated container where the power may have failed.


We can see in the trends that the energy consumption - the power went out for a little bit, the temperature may have spiked while that power is off. And what we show in these summary tiles is we can see in this temperature tile here that we've got the maximum value that’s gone out of range. Let's take a look at another display I've created which heavily uses AnalytiX-BI This is the cranes overview display. And it gives you an overview of all the RTG cranes in a port, and lets you gain some insights into which cranes may be performing or underperforming. So, the top half of this display, now we see summary charts on the various crane metrics or KPIs. And we can also filter according to the entire port or what terminal. So, let's say that we take a look at our active time across all the different grains in the port. And we can see that this grain here is got a low active time. So that's where we're going to the bottom half, that is the display. And this lets you drill down into an individual gain. So, let's take a look at the active time. And here we can see that this crane had some downtime. That's why the monthly aggregate is showing low display like this which is very useful to help you try and find any correlations in your data. For example, let's look into some of the other crane KPIs so that we can look into the stack occupancy, and we can see there is a crane that has got a much higher a full stack occupancy. And because it's six, we can then compare it to some of the other cranes. And maybe we look at the active time as well. So, the percentage of time that crane is busy. And if we compare against some of the other example cranes, we can see as the stack is fuller, it's also affecting the active time. Another application of AnalytiX-BI that I've used in this demo is for the analysis of the various crane alarms. So in this display here, you can see all the different alarms from all the different cranes in the in the port, there's a setup in the new Hyper Alarm Server, and we've stamped each alarm with some metadata like to what part of the crane what category that the alarm belongs, and then we can pull that in using the analytics data flow into it into a model and then we can sort by a terminal and by category and we can display some we can display charts like this. You can see the severity of alarms. This is like the total count of severity of alarms. Break down the alarms by shift. So, when an alarm comes in, it is just a timestamp. But using AnalytiX-BI I, you assign a timestamp to another like I read the timestamp and compare it to another table which will have the following shift, shift A B and C but that's how we modify some of the original data. This alarm data set can also be viewed down at the actual crane level. So, if I go into an individual crane, I’m viewing the site and I go into the alarm analysis of that particular crane and using that same BI data model and just adding a filter for the filter for this particular grain number. And now this chart has just been updated showing this specific count of the alarms for the T 1001 Crane.

[10:19] Mark Hepburn 

Super. Hey Alex, could you show us a little bit of the configuration of that? Would that be alright?


Thanks, Mark. That's a great question. Let me go through some of the setup, or the analytics of how I did it for the tank models. So, if you see I'm in Workbench here which is our main configuration environment. And if I go into my analytics node, we've got the BI server. And then with the way BI works: it's split into two parts. We've got data flows and data models. So, data flows is what we use to pull in the data to tidy it up. So, if I go into my tank farm example. So, for the Customers table for this demo, I just used the demo Northwind database and took the customers from there. So, the data flow lets you set up steps. So, I've got a step to import the raw table. I remove some of the columns that I don't actually need. And here I've added an email address column. So, I take the contact name and make it @company. com. So, you get a preview of the output data set. So here I've now created an email column. I do some other tidying up. I rename, change the primary key type, from one data type to another. You can do all this within the data flows. So, once you've got a data flow setup, you create a data models, so for my ports, like tanks. So, a data model is a collection of data tables and some views. So, the Customers table, I just import and input it into the data flow I just recreated. And here it shows you that the columns that we created, including the email one, I import the tank data and the tank log data. And a data model, lets you see, graphically, like a diagram of all the different tables and you add the links between the primary and foreign keys or you add the relationship links between the tables. So here, I'm linking the tank ID to the tank ID and the data log. And the customers ID to the customer ID to the data log. So, this lets you then create a data view. So, this is my data view that I'm actually using in the displays. And I'm querying the various tables that are in the same model. So, I'm getting this data set out. And you'll notice that this lets you simplify; I'm just using a SELECT query here. I don't need to do any of the joins because that is all done. Analytics-BI is clever enough to know the relationships because we've defined them in this diagram here. And then we allow you to simplify your query here. And then this data view is stored within memory and it's fast and it's cached. And that's what I'm pointing to in my GraphWorX displays. So that's a little bit of the background of how the AnalytiX-BI has been set up for that tank farm example. Back to you.


Super a thank you very much Alex. Alex, Senior Applications Consultant down in Melbourne, Australia.