Category: Odat
Author: Nanno Scheringa

Why data analysis on your logistics data pays off

Problem description

As a logistics organisation that handles transports, loading times can increase as the number of transports increases. How do you deal with this and what is the potential for improvement? Because the number of transports per FTE is often according to the set standards, it often remains unclear what causes the increase. And process improvements are sometimes implemented intuitively without process analysis, thorough research or data analysis.

But do you take the bull by the horns, or will you soon run into the same problem/issue again? And then go ahead and optimize this process and implement improvements without a thorough analysis. But there's Odat, the Oliver IT Data department with this platform and services in the field of data, including the data analysis component. This enables us to provide the right insights based on your data that are needed to actually take the bull by the horns and put our finger on the sore spot(s) at the various measurement points within your logistics process.

Solution

Our Data approach (Odat) combined with our years of knowledge of logistics processes and ditto ICT solutions enable us to quickly get to the heart of the problem and thus the solution.

In order to find an explanation, we work with the customer to collect the relevant data. This relevant data is then loaded and stored in our Odat platform. Using data analysis, we then search the data for statistical correlations. For example, by investigating the relationship between the number of transports and the loading time and translating this into an algorithm. Using the algorithm from this data analysis, we can predict the loading time based on the number of expected transports. The first step has been taken!

For many of the other transport characteristics, e.g. type of truck, there is often a similar correlation with the loading time. If there is a correlation, there need not always be cause and effect.

The proportion of the number of transports with a certain characteristic can remain the same compared to the total number of transports. If this is the case, then these characteristics in themselves are not a cause for the growth of the loading time, but the number of transports with this characteristic itself grows along with the number of transports. Through this approach, we gradually arrive at an enlightening insight, with which intuitive assumptions can be refuted in the search for process improvement(s).

However, if it can be demonstrated that a characteristic leads to an increase in loading time, we have an initial starting point here and will analyse this further. In this way we peel off the data and get to the core of the problem through the analysis.

But the transport characteristics alone are not there yet, because logistics on a plant is and remains largely human work.

Business Case

For example, we had a case where an analysis showed that higher numbers of transports required structural overtime and the number of FTEs increased more than the number of employees. So there's an issue there and the employees will have to work longer to handle all transports, with the result that the transports will also have to wait longer. By calculating the number of FTEs and not looking at the available capacity, the norm of the number of transports per FTE gave a misleading picture.

The French fries bakers model

To make the effect clear, we came up with a French fries baker's model that illustrates this problem. By the way, this model also works with fries:)

In the model there is a french fries room with a french fries baker who bakes fries for a number of customers in a period of time. The frietbakker is not very efficient and only bakes 1 portion of fries at a time.

The model has 2 parameters:

  • How many customers are present at the start time and
  • The number of minutes between customers.

The number of minutes before baking 1 portion of fries is a constant. The time that the fryer is normally open is also constant and is called the FFU (FryFrying Unit).

The organization can play with these parameters to see what the effects are on the number of customers, the number of FFU needed and the average waiting time.

If the time between the arrival of the customers is less than the time before the fries are baked, a waiting time is created. The fryer therefore has to work longer than 1 FFU in order to supply all customers with fries.

However, if you calculate the number of fries per FFU for the different scenarios, this remains the same for the different scenarios.

Our model and the results of the data analysis will further help your organization to achieve the desired process improvements and efficiency. The situation described above is one of many with which we can help you with the deployment on your data and data analysis.

If you are triggered and want to know more, don't hesitate to contact us to do an Odat Quick-scan on your logistic data together with Oliver IT.

Specific questions?

Enter your e-mail address and Oliver IT will contact you as soon as possible!

Done! We will contact you soon!
Gerben Moerland Partner
Gerben Moerland