Most frequent element of a dimension by another dimension

[arnaldocaivano]

Hello,

suppose you have this simple data:

Order:
load 
OrderNumber,City,Article
inline
[
    OrderNumber,City,Article
    1,'Madrid','Washing machine'
    2,'Barcelona','Washing machine'
    3,'Sevilla','Printer'
    4,'Sevilla','Printer'
    5,'Sevilla','Television'
    6,'Malaga','Fridge'
    7,'Malaga','Oven'
    8,'Malaga','Oven'
    9,'Malaga','Oven'
    9,'Madrid','Fridge'
    10,'Barcelona','Oven'
    11,'Barcelona','Washing machine'
    12,'Barcelona','Dish washer'
    13,'Barcelona','Fridge'
    14,'Madrid','Fridge'
    15,'Madrid','Printer'
    16,'Sevilla','Oven'
    17,'Malaga','Air conditioning'
    18,'Sevilla','Air conditioning'
    19,'Sevilla','Television'
    20,'Madrid','Television'
];

I would like to obtain this output: 

I wonder what is in Qlik-Sense the easiest, most elegant, way to calculate the most ordered article(s) with chart formula(s)

In case of more articles with the same number of orders, they should be joined with concat

Please note, I am not interested in solutions with calculations within the loading script because I would like to use the formula for  whatever pair of dimensions in a more complex cube