Data-Driven Analysis for the Operation Status of the e-Commerce

Platform Based on Olist

Zhenyu Xu

1†

, Yimin Tan

2†

, Xinxue Wang

3†

and Xiang Li

4†

School of Chemistry and Materials, Jinan University, acronyms acceptable, Guangzhou, Guangdong, China

College of Science, University of Shanghai for Science and Technology, Shanghai, China

School of Mathematics and Statistics, Weinan Normal University, Weinan, Shanxi, China

College of Software, Henan University, Kaifeng, Henan, China

†

These authors contributed equally.

Keywords: Business Analysis, e-commerce Platform, Olist, Sales Condition, Customer Feedback.

Abstract: Based on the Olist sales data and user feedback from 2017 to 2018, this paper studies the sales situation of

Olist, and discusses how enterprises can further stimulate sales, strengthen cooperation between the platform

and merchants, and optimize the customer experience. The main method of this study is to use Python to

visualize the data and to conduct statistical analysis on the data. Through the analysis, we find that there is a

lack of customer loyalty for Olist. Platform services, cooperation with suppliers and product quality control

also should be improved. The products and services provided by the platform have not reached the level of

consumer satisfaction. The time difference between the platform and the supplier also exists. In this paper,

specific suggestions are given to solve the existing problems of Olist. These suggestions can also be applied

to other e-commerce platforms with similar operating conditions as Olist to increase sales and customer

satisfaction.

1 INTRODUCTION

With the development of business, statistics on

business results have become a very important part of

business activities. Any business can gain much profit

through scientific data analysis. For example,

enterprises can find various problems in the

development of enterprises. For managers, the scary

thing is not the emergence of a problem or crisis, but

the lack of a solution or control the various data of

products within a reasonable range, understand the

relationship between supply and demand, know which

products are popular and which products are not

profitable, and adjust business ideas in time to allow

enterprises to obtain more valuable information.

Business analysis and data science can be combined.

Although there are several definitions for the

concept of business analysis, it is difficult to define it

as a set of numerical tools, methodology, business

processes, and analytical modeling methods that allow

data-driven decision-making in modern companies

and organizations. To this end, it makes extensive use

of data and information science.

In this case, we collected some data from Olist,

which is the largest department store in Brazilian

marketplaces. Olist connects small businesses from all

over Brazil to channels without hassle and with a

single contract. Since we have seen the boss's needs to

process daily operational data on the Internet, we are

ready to use our capabilities to solve these problems,

and help them to make final decisions.

With the e-commerce platform’s development,

many pieces of literature studied different business

models. There are many factors that affect the

customer’s purchase. Kwahk and Kim (Kwahk & Kim

2017) analyze results show that social interaction ties

have significant positive effects on social impact

transfer factors and trust in online vendors, whereas

they do not directly influence visit intention. Social

media commitment plays a crucial role in increasing

social impact transfer factors and e-commerce

outcomes. Flanagin et al (Flanagin, Metzger, Pure,

Markov& Hartsell 2014) suggest that despite valuing

the web and ratings as sources of commercial

information, people use rating information

suboptimally by potentially privileging small numbers

of ratings that could be idiosyncratic. And the product

Xu, Z., Tan, Y., Wang, X. and Li, X.

Data-Driven Analysis for the Operation Status of the e-Commerce Platform based on Olist.

DOI: 10.5220/0011836600003607

In Proceedings of the 1st International Conference on Public Management, Digital Economy and Internet Technology (ICPDI 2022), pages 859-867

ISBN: 978-989-758-620-0

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

859

quality is shown to mediate the relationship between

user ratings and purchase intention. Buttle et al.

(Buttle 1997) carried out mail surveys on 4,250

certificated organizations; 1,220 (28.7 percent)

responded. Marketing considerations were secondary

in seeking registration, and outcomes related to

profitability and process improvement were more

highly valued than marketing benefits.

Logistics and transportation are also important for

e-commerce companies. Speranza et al. (Speranza

2018) propose that problems in transportation and

logistics had to be tackled long before computers and

Operational Research (OR) became available to

support decision making. After the first optimization

models were developed, OR has substantially

contributed to making transportation systems efficient

and companies with complex transportation and

logistics problems competitive. Gunasekaran and

Ngai et al. (Gunasekaran & Ngai 2003) analyze the

company, a small third‐party logistics (3PL) company

in Hong Kong, which has been successful in its overall

business performance and satisfying customers. This

company's strategic alliances with both clients and

customers have helped to improve the utilization of its

resources, such as warehouse space and transportation

fleets. Also, the company is in the process of

expanding its operations across greater China, to

become a full‐fledged 3PL company.

E-commerce is becoming more popular. Sarkis

and Talluri (Sarkis, Meade & Talluri 2004) realize that

a strong supporting logistics or electronic logistics (e‐

logistics) function is an important organizational

offering from both the commercial and the consumer

perspective. The implications of e‐logistics models

and practices cover the forward and reverse logistics

functions of organizations. They also have a direct and

profound impact from an environmental perspective.

Barnes et al. (Barnes 2002) analyze that fueled by the

increasing saturation of mobile technology, such as

phones and personal digital assistants (PDAs), m-

commerce promises to inject considerable change into

the way certain activities are conducted. Equipped

with micro-browsers and other mobile applications,

the new range of mobile technologies offers the

Internet ‘in your pocket’ for which the consumer

possibilities are endless, including banking, booking

or buying tickets, shopping, and real-time news.

Bichler and Zhao (Bichler, Segev & Zhao 1998)

component-based e-commerce technology is a recent

trend towards resolving the e-commerce challenge at

both system and application levels. Instead of

delivering a system as a prepacked monolith system

containing any conceivable feature, component-based

systems consist of a lightweight kernel to which new

features can be added in the form of components.

Jing Dong is one of the giants in the Chinese

market. Li Mei and Guo Chen (Li & Guo 2015)

research the issue of self- logistics in Jing Dong under

the background of the rapid development of e-

commerce. The analysis of the advantages and

disadvantages and the comparison of the third-party

logistics proved that self- logistics enjoyed significant

advantages. However, there exist potential risks of too

large investment, difficult management of staff, and

distraction from core business. Sun and Zhou (Sun,

Liu, Higgs & Zhou 2017) consider the research status

of the Internet of Things as applied to supply chain

management. A supply chain analysis model is built

under the Internet of Things environment using Jing

Dong Mall as a case study. The results show that the

Internet of Things improved the quality of information

available to Jing Dong, enhanced its management

efficiency, and improved customers' satisfaction; it

also reduced the cost of supply chain management

whilst creating more new value.

In this paper, we mainly analyze the e-commerce

orders of Olist Store in Brazil from 2017 to 2018

focused on four dimensions: monthly sales, order

confirmation time, scores of reviews, and the number

of regular customers.

Firstly, we can see that sale volume has been

increasing throughout 2017 and reached the highest

level of the year in November. The decline was

significant in December. Sales have declined every

month since 2018, but overall sales are higher than

2017. Due to the last few months of 2018 being

missing, we do not know how the sales are going in

the rest of the months. Therefore, we guess it will hit

a new high. Secondly, we analyze the order time. We

know that the confirmation time for each quarter was

less than 5 minutes, but similar to the previous

analysis, due to the autumn and winter demand being

high, the confirmation time for spring and summer

was higher than autumn and winter. Thirdly, due to the

user evaluation, it is one of the indicators used to

evaluate the quality of goods, so we analyze the user

evaluation. 59% of users gave a 5 score and 13% gave

a 1-2 score. Overall, customer satisfaction at Olist

stores is not very high. As a result, the total number of

user reviews dropped significantly in August 2017.

We speculate that the reason influencing this factor is

product quality or platform service. Because the

repurchase rate is the key to the sustainable operation

of Olist stores. In the end, we analyze the number of

regular customers. 90% of customers buy once a

month, and less than 13% of customers buy more than

three times a month. For this purpose, the Olist Store

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

860

needs to think about how to increase the short-time

repurchase rate.

By analyzing we discuss the following ways: (1)

We need to improve the efficiency of order

confirmation in spring and summer; (2) it is suggested

that Olist needs to strengthen its quality control; (3) it

is suggested that Olist may collect customer's email

and other information, and hold activities such as

recommending and promoting products.

In this article, we first explained the background of

the data. we will first conduct business analysis, and

then combine customer data to analyze customer

behavior on the basis of business analysis. Finally, we

will summarize all the above analyses, draw our own

conclusions and corresponding recommendations.

2 DATA BACKGROUND

In this paper, we used the data as a Brazilian e-

commerce public dataset of orders made at Olist

Store. The dataset has information of 100k orders

from 2017 to 2018 made at multiple marketplaces in

Brazil. Its features allow viewing order from multiple

dimensions: from order status, price, payment, and

freight performance to customer location, product

attributes and finally reviews written by customers.

We also released a geography dataset that relates

Brazilian zip codes to let/lng coordinates. Next, we

mainly study four aspects, such as monthly sales,

order confirmation time, scores of reviews, and the

number of regular customers.

In this picture, we can see that there are four

variables that affect orders. These four factors are

monthly sales, order confirmation time, scores of

reviews, and the number of regular customers. They

restrict each other. And these factors also have their

own restricting factors. The reason for affecting order

confirmation time is a customer's location, freight

performance and order statues. Monthly sales and the

number of regular customers have the same restricted

factors. It is about price and product attributes. Scores

of reviews have only affected reasons for reviews

written by customers.

3 BUSINESS ANALYSIS

In this section, we selected some data from the

homepage of Olist. After that, we used these data to

do sales analysis, delivery time analysis and order

time analysis.

3.1 Sales Analysis

We first created a table to look at the total sales of

each month. We found that each month's sales were

increasing. However, due to the missing data in the

last few months of 2018, we were unable to make a

judgment. As we can see from the Table 1, the

maximum sale is in November, which is 7544.Then

we take a graph for the sales amount from 2017 to

2018.

From the total sales (Figure 1), it can be seen that

regardless of the lack of data, the sales volume showed

Figure 1: Variable diagram.

Data-Driven Analysis for the Operation Status of the e-Commerce Platform based on Olist

861

Table 1: Total sales for each month in 2018 and 2017.

Month Jan Feb Ma

May Jun

2017 800 1780 2682 2404 3700 3245

2018 7269 6728 7211 6939 6873 6167

Month Jul Aug Sep Oct Nov Dec

2017 4028 4331 4285 4631 7544 5673

2018 6292 6512

Figure 2: Total sales from 2017 to 2018 sales.

Table 2: Delivery Time Information (Excerpt).

Customer date Deliver

date Season

1 2017-10-10 2017-10-18 autumn

2 2018-08-07 2018-08-13 summer

3 2018-08-17 2018-09-04 summer

4 2017-12-02 2017-12-15 autumn

5 2018-02-16 2018-02-26 winter

Time difference Difference float

1 8days+21:25:13 -7.107488

2 6days+15:27:45 -5.355729

3 18days+18:06:29 -17.245498

4 13days+00:28:42 -12.980069

5 10days+18:17:02 -9.238171

a rapid upward trend in 2017 and reached its peak in

November. After that, it has remained stable and

fluctuates at around 6,500. It can be seen from this that

2017 was the peak period of Olist’s development, and

the order volume increased sharply. By 2018, the sales

volume stabilized and the development entered the

stabilizer. Data to speculate how to design marketing

strategies.

3.2 Delivery Time Analysis

We counted the logistics speed of each season. The

method of analysis is to make the difference between

the estimated delivery date provided by the store and

the actual delivery date, and finally get the difference

date. The time difference column is obtained by

subtracting the second column (delivery date) from

the first column (customer date), the last column

(difference float) is the number that turns the date into

decimal Part of the whole table is shown in table 2.

Afterward, we plotted the difference (Figure 2) in

days in different seasons and found that no matter

what season, the customer's delivery time was much

earlier than the expected delivery time, about a week

earlier on average. According to the distribution of the

histogram, the logistics rate in autumn and winter is

significantly higher than that in spring and summer.

The reason may be that the demand in spring and

summer is larger, while the demand in autumn and

winter is less.

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

862

Figure 3: Time difference between estimated and actual delivery time.

Table 3: Order Time Information (Excerpt).

Order

urchase timestam

Order a

roved at

1 2017-10-02 10:56:33 2017-10-02 11:07:15

2 2018-07-24 20:41:37 2018-07-26 03:24:27

3 2018-08-08 08:38:49 2018-08-08 08:55:23

4 2017-11-18 19:28:06 2017-11-18 19:45:59

5 2018-02-13 21:18:39 2018-02-13 22:20:29

season Time difference Time difference float

1 autumn 0 days 00:10:42 0.007431

2 summe

1 da

s 06:42:50 1.279745

3 summe

0 days 00:16:34 0.011505

4 autumn 0 da

s 00:17:53 0.012419

5 winte

0 days 01:01:50 0.042940

3.3 Order Time Analysis

We then compared the difference between the

payment time and the merchant's approval time in

different seasons. The time difference column is

obtained by subtracting the second column (delivery

date) from the first column (customer date), the last

column (time difference float) is the number that turns

the date into decimal.

According to Figure 4, we found that the approval

time for each season is within five minutes. Similar to

the project analyzed in the previous analysis, the

approval time for the spring and summer seasons is

higher than that for the autumn and winter seasons.

The reason may also be due to the large number of

orders in the spring and summer seasons. In the

autumn and winter seasons, more than 80% of the

response time was within one day. In the spring and

summer seasons, due to the large sample size and

network delays caused by many orders, the response

time within one day dropped to 70%.

Through the above analysis, we can summarize

that it can be seen from the sales volume that the

overall Olist is still developing, but the speed of

development is slowing down. This is a problem

worth thinking about, and a strategy needs to be

devised to solve this problem. Turning our attention to

logistics and response time, it can be seen that

although the overall is still within an acceptable time

range, there is still a lot of room for improvement,

because, with the development of the times, people's

tolerance for waiting time will decline, so a method to

increase the speed of logistics is very necessary.

4 CUSTOMER’S ANALYSIS

In this part, through the analysis of customer’s review

scores and the change of customers repurchases. We

can figure out how the public praise of Olist’s products

changed and give Olist the corresponding future

development plans.

Data-Driven Analysis for the Operation Status of the e-Commerce Platform based on Olist

863

Figure 4: Order Validation Time.

Table 4: Customer’s Review Scores.

month score Review numbers

1 January 1-2 1032

2 January 3-4 1806

3 January 5 3695

4 February 1-2 1170

5 February 3-4 2114

Figure 5: Customer’s Review Scores.

4.1 Score Evaluation

In order to observe customers’ response to the project,

according to the Olist’s data source, we counted the

number of costumer’s review scores and made a

corresponding table. In the Table 4, we can find that

no matter what month it is, the number of customers

who give 5 points is the largest, then followed by

customers who give 3 to 4 points. And customers who

give 1 to 2 scores are the least.

It is widely known that the scores given by

customers have a great relationship with the quality of

goods and clearly reflects the feelings of customers

after they purchase commodities. Besides, users’

review rating is the most critical index to evaluate the

quality of goods and services provided by e-commerce

platforms.

We can see clearly in the Figure 5 that Olist's

overall customer satisfaction is not satisfactory. Only

59% of customers gave a rating of 5 points out of 5

points. 13% of users gave negative reviews of one or

two points. In the first eight months, we can see a

steady change in the data. In August, the number of

five-point positive reviews suddenly dropped sharply.

Then it became smooth after September and then

steadily increased again after November.

In August, there was an unusual drop in users’

ratings. Such an abnormal situation may be due to

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

864

large-scale quality problems of the products being

sold, or serious service errors of the platform.

Considering the abnormal decline of both sales and

reviews in August, it is suggested that Olist should

strengthen its quality control. For Olist, ensuring the

quality of products should be the top priority. Only by

improving the quality of goods can customers'

satisfaction after purchase be increased.

4.2 Regular Customers

At last, we counted the number of the customers'

repurchases. Attracting customers to spend repeatedly

on the platform is the key point of the sustainable

business of e-commerce because whether it can attract

customers to repurchase is also an important indicator

of the electronic commercial success.

In order to observe the changes of the data more

easily, we still make the data into a line chart. In the

chart, the blue line means customer buy this product

for the first time, the orange line means customer buy

it the second time and the green line means customer

purchases three or more times.

According to the Figure 6, we can draw the

conclusion that before August, the volume of

customers' first purchase increased steadily at a slow

and steady pace. However, in August there was a

sharp decline due to some specific reasons. After

September, there has been a significant rebound in the

amount of data of customer purchase. Meanwhile, we

can find a problem that during the whole period of the

year, the number of second and third purchases

remained steadily at a lower level above 1,000,

without drastic changes, and even did not change with

the data of first purchases.

According to the data of regular customers, Olist

had a bad performance in getting customers to make

orders again. In 2017, more than 90% of Olist's users

placed only one order per month, and less than 3% of

customers spent three or more orders per month on

Olist. Olist needs to think about how to motivate users

to make multiple purchases in a short period of time.

According to the marketing strategy, it is suggested

that Olist may collect customers' email and other

information, recommend products and promotion

activities regularly, and develop different activities

according to different themes throughout the year, so

as to attract customers to repeat shopping on the

platform within a short period of time.

By comparing the number of customer’s first

purchases and the scores given by customers. We can

find that they were inextricably linked. The trend was

roughly the same between them both. It suggesting

that there is a huge connection between the customers'

purchase and the public praise of the products.

According to the above conclusion, we suggest

that Olist should increase the logistics speed in the

spring and summer seasons because this is greatly

related to customer satisfaction and the time from

order to confirmation can be maintained cause this

time is within an acceptable range. Otherwise, Olist is

not doing well in terms of the repeat customers, they

should take appropriate actions to recall regular

customers, such as discounts on products and so on, to

attract customers to repeat purchases. It also can be

seen from the discount chart that the quality of

products in August and September may decline, which

leads to a decline in repurchases. It is necessary to

strengthen the quality control to make customers most

satisfied. And the public praise of products will also

improve accordingly.

Figure 6: Customers repurchases.

Data-Driven Analysis for the Operation Status of the e-Commerce Platform based on Olist

865

5 DISCUSSION

5.1 Extension

Due to the lack of some data in this study, there is a

certain deviation between the actual sales situation of

Olist and the analysis and prediction. In the absence of

valid review content data, it is impossible to pinpoint

the reason for the sudden drop in ratings in August

2017. Further customer feedback data needs to be

collected and its trend needs to be analyzed to

determine whether Olist's products and services were

incorrectly provided, or whether competitors had

produced a large number of malicious low ratings.

For sales, two years’ data was not enough to make

accurate predictions. With the growing popularity of

online consumption, Olist's sales are likely to grow

even more sharply after 2018. To judge whether Olist

is competitive among enterprises of the same type at

the same time, more data of competing businesses and

itself are needed for comparative analysis before more

accurate and effective suggestions can be given.

5.2 Recommendation

Considering the operating condition of Olist, we have

three recommendations:

(1) To improve the order validation efficiency,

Olist should strengthen the contact with cooperative

merchants and shorten the time of order processing. It

is suggested that Olist introduce a policy that requires

general merchants to process orders within 24 hours,

and merchants selling customized goods should also

process orders within 48 hours, so as to ensure

consumers' shopping efficiency.

(2) To handle the problems with product quality,

Olist should strengthen its quality control. Olist can

also require suppliers to pay customers the double

value of the product and a certain amount of penalty to

the platform when quality problems emerge.

(3) To attract Regular customers, Olist may collect

customers' email and other information to recommend

products and promotion activities regularly. Also,

Olist should develop different activities according to

different themes throughout the year, so as to attract

customers to repeat shopping on the platform within a

short period of time.

6 CONCLUSION

Olist is a Brazilian e-commerce platform. According

to more than 10,000 pieces of public sales data from

2017 to 2018, the marketing status and existing

problems of Olist were analyzed in terms of sales,

service condition and customers’ satisfaction. In this

article, we used Python to visualize the data, and we

can clearly see the annual sales trend, the distribution

of order confirmation time, the change of customer

rating, and the distribution of consumption frequency.

First, based on the annual sales trend, we get into

the result that the holiday effects such as Thanksgiving

Day and Christmas have a significant stimulus to the

sales of Olist. The sales trend in the first half of the

year was slightly weaker than in the second half due to

the lack of effective motivation. Second, the time of

order confirmation and delivery efficiency indicates

that Olist, as an e-commerce platform, still has

problems in cooperation with suppliers. Although the

overall order confirmation time is less than 24 hours,

there are still some orders that take more than a day to

confirm. As a platform, consumers' shopping

experience should be more taken into account, and

time should be reduced to improve order efficiency.

The last point is that customer rating and consumption

frequency are inseparable indicators of the two boxes.

According to the data, a considerable number of

customers are not satisfied with Olist's services or

products, and only a small number of consumers are

accustomed to shopping on Olist. We suggested that

Olist introduce relevant policies to improve the quality

control of platform services and suppliers' products. At

the same time, increased marketing efforts to promote

the penetration of online shopping in Olist consumer

life habits.

While writing this thesis, we realized that there are

still many deficiencies as follows: (i) The whole

analysis of this paper is based on the data of Olist. So,

the data analyzed has great limitations. Besides, the

data is only limited to one year's sales, which cannot

possess enough universal. The resulting data changes

are also applicable to Olist and cannot cover all aspects

of online sales. (ii) The commercial data analysis

based on Olist makes us lack some innovation and it is

difficult for us to innovate the data analysis that

belongs to our own creation. (iii) The limitation of data

makes the number of references we can use is too

small and not authoritative enough.

Based on the defects in the above data analysis, we

need to improve our analysis framework in future

studies. firstly, we could find out reasons for data

changes more clearly and enhance the accuracy of data

analysis. Then, the data sources we analyze should

also include broader aspects to enhance the

universality of our data analysis. Last but not least, we

should also add some innovative data analysis

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

866

methods in the article to help us make an analysis, so

that we can draw some more valuable conclusions.

ACKNOWLEDGEMENT

First and foremost, we would like to express our

sincere gratitude to Professor Noah Gift and mentor

Amandeep from University of California Berkeley,

who let us into the world of data analysis. Professor

Gift used his exquisite technology to show us the

charm of data analysis, which aroused our interest in

data analysis. Mentor Amandeep carefully answered

our questions during the learning process and offered

constructive suggestions and guidance for the

completion of the project.

Secondly, we would like to express our gratitude

to Mr. Yuxiang Cheng, our project supervisor, who

was always patient during the paper-writing process

and gave us great help. Before starting the paper, he

explained to us about the structure of paper in detail.

And he pointed out all of the matters needed to be

corrected during lessons. We were deeply touched by

his patience. Also, we would like to thank Ms. Sun Ke

for her valuable advice on our final stage. These

suggestions make our paper better.

Last but not least, we would like to express our

gratitude to our family and friends for their support

and understanding during our study and writing.

Thanks again to everyone who has been giving us

guidance and help. We will continue to explore and

advance in the academic field.

REFERENCES

Buttle, F. (1997). ISO 9000: marketing motivations and

benefits. International journal of quality & reliability

management.

Barnes, S. J. (2002). The mobile commerce value chain:

analysis and future developments. International journal

of information management, 22(2), 91-108.

Bichler, M., Segev, A., & Zhao, J. L. (1998). Component-

based e-commerce: assessment of current practices and

future directions. ACM SIGMOD Record, 27(4), 7-14.

Flanagin, A. J., Metzger, M. J., Pure, R., Markov, A., &

Hartsell, E. (2014). Mitigating risk in ecommerce

transactions: perceptions of information credibility and

the role of user-generated ratings in product quality and

purchase intention. Electronic Commerce Research,

14(1), 1-23.

Gunasekaran, A., & Ngai, E. W. (2003). The successful

management of a small logistics company. International

Journal of Physical Distribution & Logistics

Management.

Kwahk, K. Y., & Kim, B. (2017). Effects of social media

on consumers’ purchase decisions: evidence from

Taobao. Service Business, 11(4), 803-829.

Li-mei, Q. I., & Guo, C. H. E. N. (2015). Research on Self-

Logistics and Distribution System of Jingdong Mall.

Journal of Shandong Institute of Commerce and

Technology, 02.

Speranza, M. G. (2018). Trends in transportation and

logistics. European Journal of Operational Research,

264(3), 830-836.

Sarkis, J., Meade, L. M., & Talluri, S. (2004). E‐logistics

and the natural environment. Supply Chain

Management: An International Journal.

Sun, L., Liu, J., Higgs, R., & Zhou, L. (2017, July). Analysis

on the Application of the Internet of Things Technology

to Jingdong Mall's Supply Chain Management. In 2017

IEEE International Conference on Computational

Science and Engineering (CSE) and IEEE International

Conference on Embedded and Ubiquitous Computing

(EUC) (Vol. 2, pp. 437-442). IEEE.

Data-Driven Analysis for the Operation Status of the e-Commerce Platform based on Olist

867