Use of Data Analytics to Increase the Efficiency of Last Mile Logistics for Ecommerce Deliveries

Use of Data Analytics to Increase the Efficiency of Last Mile Logistics for Ecommerce Deliveries

Gaurav Nagpal (Birla Institute of Technology and Science, Pilani, India), Gaurav Kumar Bishnoi (Birla Institute of Technology and Science, Pilani, India), Harman Singh Dhami (Birla Institute of Technology and Science, Pilani, India) and Akshat Vijayvargia (Birla Institute of Technology and Science, Pilani, India)
DOI: 10.4018/978-1-7998-3053-5.ch009
OnDemand PDF Download:
No Current Special Offers


With the increasing share of digital transactions in the business, the way of operating the businesses has changed drastically, leading to an immense opportunity for achieving the operational excellence in the digital transactions. This chapter focusses on the ways of using data science to improve the operational efficiency of the last mile leg in the delivery shipments for e-commerce. Some of these avenues are predicting the attrition of field executives, identification of fake delivery attempts, reduction of mis-routing, identification of bad addresses, more effective resolution of weight disputes with the clients, reverse geo-coding for locality mapping, etc. The chapter also discusses the caution to be exercised in the use of data science, and the flip side of trying to quantify and dissect the phenomenon that is so complex and subjective in nature.
Chapter Preview

Literature Review

The optimization of logistics decisions has become indispensable with the increased movement of goods and services across the geographies in globalized trade (Langevin and Riopel, 2005). Koul and Verma (2011) used advanced analytics to consider the influence of the uncertainties tied to the human cognitive thinking process for vendor selection. Similar studies on the power of analytics in revolutionizing supply chain have been done by LaValle et al. (2011), Chen et al. (2012) and Khan (2013). Waller and Fawcett (2013) said that supply chain will be revolutionized and transformed by predictive analytics to improve the productivity and operational efficiencies. Wu et al. (2016) while doing the review of smart supply chain literature also suggested that there is a tremendous potential on the research for applying analytics in the supply chains. Hanne and Dornberger (2017) also presented how analytics can be used for transportation planning and vehicle routing problems. Deep et al. (2019) presented many of the latest innovations and analytics applications for supply chain, inventory and logistics.

Key Terms in this Chapter

Classification Table: A table that captures the predicted number of successes (or failures) to the observed number of successes (or failures).

Predictive Modeling: Application of statistics and data to make prediction of the outcomes using data modelling.

Feature Engineering: Application of domain knowhow to pull out features from raw data with the use of data mining.

Route Optimization: Finding the least costly or the fastest route for a set of shipments with complicated constraints related to customer time windows, driver availability, vehicle availability, traffic situations, road conditions, etc.

Confusion Matrix: A table that captures the performance of a classification algorithm on a dataset for which the observed values are known.

Complete Chapter List

Search this Book: