A Performance and Efficiency Analysis of Logistic Data

In this analysis, we will delve into nine years of logistics data encompassing key variables such as location, date, and price. By examining these factors over an extended period, we aim to uncover patterns and trends that could offer valuable insights into the logistics processes. Our goal is to identify opportunities for optimization and enhance overall efficiency within the supply chain.

When we look at the distribution of the number of orders over the past 9 years, we see that there is not much change or development, he price trend seems to continue around 240. We can see the increase in orders, albeit slightly, with the price decrease in 2018. In addition, there is an increase in prices and a decrease in the number of orders in 2019.

For the monthly distribution, we can see that although there is not much order change between months, but March and May are higher than other months.

The reason for the increase in orders in May and July may be price decreases.

However, in order to obtain accurate predictions on a monthly basis, it is necessary to analyze on a yearly basis. More details will be discussed later in the study.

When we look at 2018 and 2022 on a monthly basis, we see a serious decrease in the 12th month.

When we look at monthly changes in general, trends change a lot from year to year, so I wanted to look at seasonal trends to get a more understandable interpretation. Seasonal trend graph as below.

In general, except for 2019 and 2023, we can say that there is an decreasing trend in prices in the spring seasons and an increasing in the summer and winter seasons.

When we look at the distribution of the number of orders based on location, it is very difficult to see a sharp difference. There is an equal distribution although there are changes in prices according to cities. Just as a comment, we can say that there is a price-transportation balance here. I think geographical conditions affect the price.

When we look at the effect of the days of the week on price and order, we see that they are both equally distributed on a daily basis. We can say that there is no density or price change depending on the days of the week.

Likewise, I want to see the effect of the time of day on the price and order. There is no price change on an hourly basis.

Although there is not much difference in record density, we can say that the number of records is slightly higher at 5 am and at 12 am.

Now, to get to know the data better, I want to see the yearly price distribution based on location.

According to the price distribution, we can say that San Jose, New York, San Dieago, Los Angeles have higher prices, while Phoenix, San Antonio have lower prices. This seems to be related to location. When we look generally, we see that there are outlier prices for every city.

I think this difference on the maximum side is related to the sensitive service scope like temperature controlled and may be useful in the future to make sense of the data. Likewise, I think the outliers on the lower side are related to services that do not require sensitivity. It may depends on service type.

2. Detailed Analyzing per Year by City

Firstly, there was a serious price drop in 2016. This situation is also noteworthy for 6 other cities. A political or economic global situation may have an impact this year.
The one of the things that stand out here is that both the price trend and the demand have increased in 2018. We clearly see that there is a high demand in this year.
Lastly, here is a serious decrease in demands and prices in 2020. The reason for this may be the pandemic.

We can say that it has similar characteristics to San Jose, except for the price trend in 2022.

The price trend continues to decrease until 2019, but demand and prices are increasing in 2019.

We can say that there has been a serious decrease in the price trend after 2017 and there has been an increase in orders with this low price trend until 2021. However, although the price trend did not change after 2021, the number of orders decreased again. From here we can say that they are strong rivals for San Diego and should exhibit competitive behavior.

We can say that both prices and orders decreased for this city in 2019. This may be due to competition. And while prices were still low in 2020, orders increased and we can say that low prices started to show their effects in 2020.

For Chicago, except for 2018, we see the number of orders inversely proportional to the price trend. Therefore, it may be a price performance preference.

There was a serious decrease in orders in Houston in 2016. In 2017, there is a significant increase in orders with the price trend. Just as a guess, with a new service, it may be preferred even if the prices are slightly higher. This situation seems to have affected the price trend in Dallas because there is a serious decrease. Maybe warehouses in Houston are more likely to be preferred than Dallas because they are closer to the port.
But in 2018, the opposite situation is happening, while orders are increasing along with prices in Dallas, orders are decreasing along with prices in Houston. I think that there is a different dynamics that related to sector. Every year there seems to be a completely opposite dynamic between the two cities. This may indicate that they need to be competitive with each other. Because they are not actually very far away and the services they provide may have a big impact on their preference and price dynamics.

The price trends of San Antonio and Phoenix are very similar to each other, so we can say that they are two competitive cities. Price trends look generally similar. Prices are quite low compared to other cities. The reason for this is most likely because they are not a port city. I'm sure that for the customer, shipping costs for such regions also come into play. So, we can say that they achieved sales by keeping prices low. The sudden price increases in 2016 and 2020 for these two cities are remarkable.

Overall;

While there was a general downward trend in prices in port cities in 2016, there was a price increase trend in other cities. We can comment that it may be a major factor across the country this year, maybe it should be economy.

I think that especially San Antonio and Phoenix are oppositely affected by the situation across the country compared to other cities.

We can see that there may be many factors affecting warehousing preferences, and price is not the only criterion. So, we need more information for more insights.

3. Detailed Analyzing per Month by City

Graphs

Comments

San Jose

In general, a price decreasing trend is observed in the 2nd and 12th months.

As for orders, we see that the highest number of orders is in May, especially with the increase in 2023 and 2020.

New York

For New York, we see that there is generally a price increase in February and November, except for a few years.

We see that orders increase in January, October and November, especially in 2023.

In Los Angeles, we see the most orders in total in March, especially with the increase in orders in March 2019.

We see an increase in prices in October for 3 years in 2018, 2019 and 2020

Dallas

We cannot say that there is a consistent trend in price trends between years and months, but we can say that there is a price trend that increases in July and October and decreases in September and December in 2023.

It received its highest orders in January, with the highest orders in 2015, 2016 and 2023. In addition, there is a significant increase in May 2023.

Houston

While there was an increase in orders in Houston in January and November 2023, considering all years, the most orders were received in November.

In the year-by-year analysis above, we saw a serious decrease in orders in Houston in 2016. When we look at this situation on a monthly basis, we can see that this decrease occurred especially in June and August.

Chicago

There is an increase in orders in Chicago in 2021. We can also see the price decrease in May this year.

San Antonio

In San Antonio, we can see a serious drop in orders in July 2019 and also a drop in the price trend this year.

Phoenix

For Phoenix, the first thing that stands out is the decrease in orders in February (especially 2015) in total. When we look at the price trend, there is a significant increase in February 2015, which may be the reason for this situation.

Overall;

Considering all the years, it is a little difficult to establish a price trend on a monthly basis. But some years were similar to each other. For example, for San Jose, 2022 and 2023 show similar trends except for May and July. Or we can say that 2022 and 2023 show similar trends except for the 6th and 11th months for New York. The information here can be interpreted and used according to need and goal.

Except for Chicago, San Antonio and Phoenix; In common with other cities, we can say that the order numbers are good in May 2023. Also, there are significant increases in different months on a city basis, for example, we can say July for San Jose and San Diego.

4. Outlier Analysis

Since I thought that some high prices might be related to service (like product, condition, quality), I wanted to analyze them and add them to the data as features to make the data more descriptive.

Values labeled as high prices are taken as those greater than 80% of the price distribution within each city on a yearly basis.

A total of approximately 200 high price values were found for each cities. I think these high priced orders offer specific services such as requiring extra storage conditions.

When I look at the hourly change of high price orders, I see that there are high price orders in the evening hours in Chicago and Houston, and in the night hours in Los Angeles, NY, Phoenix, San Antonio and San Diego. Therefore, although we cannot see it very clearly, we can say that the effect of high prices may be related to the time of day.

In conclusion, our analysis of the nine-year logistics data has provided a comprehensive understanding of the underlying patterns and trends in the supply chain. By leveraging these insights, organizations can make informed decisions to optimize their logistics operations, reduce costs, and improve overall efficiency, ultimately driving better performance and customer satisfaction.

Birsen Bayat