The Machine Learning Artifact for Logistics Business Analysts – Amazon SageMaker Canvas Product Review

Facing the rising demand for data analysis. Amazon cloud technologies has launched exciting new machine learning capabilities as digitization and the accelerated adoption of artificial intelligence have created a severe shortage of artificial intelligence (ai) and machine learning (ml) developers in the enterprise. – amazon sagemaker canvas.

Amazon sagemaker canvas further lowers the barrier to entry for machine learning. Enabling operations analysts to use a point-and-click interface to make more accurate machine learning predictions without coding. This allows anyone interested in learning more about ml to try out the technology for free.

Without further ado. Let me experience the charm of sagemaker canvas with common data in the vertical subdivision of logistics !

Analysis scenarios and data preparation

The correlation between the picking quantity and the time-consuming and quality of each time is analyzed. This time. The correlation is found from the factors of the number of boxes transported within the specified time of each link. The time-consuming of each link. And the quality status of the boxes in each link.

 

Simulate the logistics operator to pick different numbers of boxes within the specified time. The final completion rate and quality are different. Analyze and find out the best number of picking boxes for a single operation. For business operation students. When formulating transportation task strategies. There is data to refer to and make more reasonable and efficient operation strategies.

It is not important whether the conclusion of the Oman Phone Number analysis is correct. What is important is to simulate the relationship + explore the analysis methods from different angles + the use of different tools.

Desensitized and virtualized warehouse intralogistics data.

Oman Phone Number
Oman Phone Number

In the logistics operation process in a warehouse of a certain factory, there is a section of “issue the task of requesting goods – sorting items – shipping items – delivering items”. The current completion rate, the time used for each delivery, and the quality of the items in each link are recorded as the following:

 

 Evaluation experience steps and data analysis structure

Register and log in an account, search for SageMaker in the console, and click to enter the SageMaker console. 2. Import and manage data Create a new bucket in S3 and import data. Note that the import format is .csv, and the document cannot contain Chinese. I did not pay attention to the format specification at the beginning of the experience, and the import failed several times. The overall experience is relatively smooth. When logging in and setting up for the first time, you need to read more guidance documents and pay attention to the page prompts. 3. Launch the application In SageMaker Domain, find Canvas (Chinese version: Canvas), click to start [Canvas]. 4. Deployment In about 1 minute, the back-end application deployment is completed.

In SageMaker Canvas, click [+New Model].

Import the data in S3 just now.

In this analysis, I chose [demand case] as the target column, and SageMaker Canvas will automatically use the appropriate model for analysis.

Why choose [demand case] as the target column, because [demand case] requires the number of containers, and all logistics operations are based on [demand case] to have subsequent process, time-consuming, and quality analysis. For another example, if your analysis is about advertising and marketing, the target column can be the first level of the marketing funnel.

The field [NO.] is a serial number, which is not related to the data. If this field is not checked, it will affect the prediction; other fields have high correlation and should be reserved.

Leave a comment

發佈留言必須填寫的電子郵件地址不會公開。