A bibliometric analysis of inventory management research based on knowledge mapping
| Pages | 127-154 |
| Published date | 04 February 2019 |
| Date | 04 February 2019 |
| DOI | https://doi.org/10.1108/EL-11-2017-0241 |
| Author | Yong Ye,Yuanqin Ge |
| Subject Matter | Information & knowledge management,Information & communications technology,Internet |
A bibliometric analysis of
inventory management research
based on knowledge mapping
Yong Ye
Department of Logistics Engineering, Anhui Agricultural University, Hefei, China, and
Yuanqin Ge
Chinese Graduate School, Panyapiwat Institute of Management,
Nonthaburi, Thailand
Abstract
Purpose –The research mainly aimsat the hotspot of inventory management by knowledgemapping and
providesa visualization referencein this research field.
Design/methodology/approach –First, inventory management journals during 1986 to 2017 were
selected as the research object and text formattingin the Web of Science (WOS) database is exported. Then
inventory management knowledge mapping is done and clustering keywords are extracted by using
CiteSpace and VOSviewer software. Based on co-word analysis, the three special clusters are exported:
inventoryoptimization strategy, inventory pricing and inventorytechnology. Besides, the clustering structure
and time evolutionare analysed. Finally, bibliographic item co-occurrencematrix builder (BICOMB) was used
to extract the “journal”and “researchers”keywords in the inventory management research fields. Setting
three parameters such as the cited half-life,centrality, frequency and keywords for data mining, it can infer
the trend keywordsof future research.
Findings –Results showed that inventorymanagement research has been abundant in literature overthe
past 30 years and has experienced a changefrom focusing on inventory optimization strategy to inventory
pricing and inventory technology in process. It shows that inventory management research focused on the
classic topics and includes economic order quantity, dynamic pricing, design and technology, and thenew
topics includechannel coordination, hierarchical price and simulation.
Research limitations/implications –Based on knowledgemapping, this study is still relativelymacro
and cannot cover all areas of inventorymanagement. This study only investigated the state of correlational
researchin WOS and Google Trends and not additional databases.
Originality/value –The currentresearch mainly builds on knowledge mappingfor the research hotspot of
inventorymanagement and provides visual references for future researchin this field.
Keywords Inventory management, Bibliometric analysis, Co-citation analysis,
Knowledge mapping
Paper type Research paper
Introduction
Inventory management has been an area of greatactivity, both for enterprise managers and
researchers. A basic search of the keyword “inventory-management”returns a total of
13,770 papers in the Web of Science(WOS) Core Collection (1986-2017), all of which have an
This work was supported by Anhui Provincial High School Provincial Quality Engineering Project
(2016jyxm0308), the National Natural Science Foundation of China (No. 31371533, No. 31771679,
No. 71771003) and the Natural Science Foundation of Anhui Province, China (No. 1808085MG215).
Inventory
management
research
127
Received23 November 2017
Revised1 April 2018
4 May2018
Accepted22 May 2018
TheElectronic Library
Vol.37 No. 1, 2019
pp. 127-154
© Emerald Publishing Limited
0264-0473
DOI 10.1108/EL-11-2017-0241
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0264-0473.htm
H-index of above 124. The earliest study in this field dates back to the first decision model
that aided managers in decisions regarding the scale and time of inventory replenishment
order; that is, the economic order quantity (EOQ) model put forward by Harris (1913).
Subsequently, in the 1950s, inventory management researchers primarily focused on
answering questions about the quantitiesof orders and the time at which they were placed.
Later, the EOQ model expanded into three important branches: the economic production
quantity (EPQ) model, the conceptof reorder point and the random EOQ model (Taft, 1918;
Whitin, 1954;Wilson,1934).
Inventory management is generally divided into either deterministic policy or random
policy. The deterministic model assumes that demand and delivery time are completely
definitized in the inventory system, which builds the model within lead time and has a
simple mathematical structure. The random inventory model was first proposed by Arrow
et al. (1951) and Dvoretzky et al. (1952a,1952b)to study the uncertainty and included service
level and order fill rate. Unlike the deterministic model, the mathematical structure of the
random model is more complicated.
Knowledge mapping is a tool that is used inscientometrics to analyse and establish the
structure or function of the knowledge field and visualize them in comprehensive and
transparent format. In order to contribute to inventory management research, this paper
uses the WOS as a data source and the newest CiteSpace and VOSviewersoftware to create
the knowledge graph, so as to provide a valuable reference for inventory management
policy.
Methods
Data sources and methodology
On 15 March 2017, “inventory management”waschosen as a phrase to search documents in
the kernel database of WOS, contraryto the study of Romo-Fernández et al. (2013). That is, a
journal was chosen and keywords were extracted to make an analysis in this journal. The
keywords were downloaded from all citable documents published in all journals, obtaining
8,040 documents after retrieving the inventory subject from two categories: “Operations
Management”and “Management”between 1986 and 2017. Therein, there was a total of
83,741 cited documentsand 33,784 citing documents excluding self-citation,with an average
citation of 16.27 times for each item. All of these documentswere fed into a text file and the
knowledge graph was constructedafter the standardization of keywords.
Keyword standardization was realizedby adopting Levenshtein’s edit distance and best
character string matching the Damerau–Levenshtein distance algorithm to distinguish
similar words. Therefore, the Levenshtein distance (Levenshtein, 1966) is the necessary
minimum edit operand in converting one-character string to another character string. The
improvement of the Damerau–Levenshteindistance (Damerau, 1964) lies in the match of the
best character string, including the transpositionbetween two characters besides insertion,
deletion and substitution. In the Levenshtein distance, the final action is regarded as two
operations, that is, deletion and insertion. Specifically, the procedure of keywords
standardizationis:
removing any punctuation with no meaning, like count mark;
uniting singular and plural; and
pairing of similar keywords based on the Levenshtein and Damerau–Levenshtein
distances.
EL
37,1
128
To continue reading
Request your trial