Logistics Innovation and Social Sustainability: How to Prevent an Artificial Divide in Human–Computer Interaction

• DOI: http://doi.org/10.1111/jbl.12198
• Author: Henk Zijm,Matthias Klumpp
• Published date: 01 September 2019
• Date: 01 September 2019

Logistics Innovation and Social Sustainability: How to Prevent an

Artificial Divide in Human–Computer Interaction

Matthias Klumpp

1,2,3

and Henk Zijm

4

1

Fraunhofer Institute for Material Flow and Logistics (IML)

2

University of G€ottingen

3

FOM University of Applied Sciences

4

University of Twente

Human–computer interaction (HCI) is a cornerstone for the success of technical innovation in the logistics and supply chain sector. As a

major part of social sustainability, this interaction is changing as artificial intelligence applications (Internet of Things, autonomous trans-

port, Physical Internet) are implemented, leading to larger machine autonomy, and hence the transition from a primary executive to a supervi-

sory role of human operators. A fundamental question concerns the level of control transferred to machines, such as autonomous vehicles and

automatic materials handling devices. Problems include a lack of human trust toward automatic decision making or an inclination to override

the system in case automated decisions are misperceived. This paper outlines a theoretical framework, describing different levels of acceptance

and trust as a key HCI element of technology innovation, and points to the possible danger of an artificial divide at both the individual and firm

level. Based upon the findings of four benchmark cases, a classification of the roles of human employees in adopting innovations is developed.

Measures at operational, tactical, and strategic level are discussed to improve HCI, more in particular the capacity of individuals and firms to

apply state-of-the-art techniques and to prevent an artificial divide, thereby increasing social sustainability.

Keywords: artificial intelligence; social sustainability; logistics performance; human-computer interaction

INTRODUCTION

Logistics and supply chain management are subject to rapid

changes as a result of technological, social, and market evolutions

within the global economy, see, for example, Bloemhof et al.

(2015), Hilger et al. (2016), Sternberg and Norrman (2017), Ber-

tazzi and Mogre (2018), or Fors et al. (2015). In response to

increasing customer demands (cost effectiveness, sustainability,

speed, tailored problem solutions), automation in production and

distribution has migrated from the execution of programmed tasks

to a level, in which software agents and robots act (partly) autono-

mously using artificial intelligence (AI)-based algorithms (Gun-

sekaran and Ngai 2014; Lee et al. 2014; LeCun et al. 2015; Torabi

et al. 2015; Kong et al. 2016; Castillo et al. 2017). A key question

that comes along with these developments concerns the future

form and performance of human-computer interaction (HCI).In

the past, working areas of robots and humans in production and

transportation were largely separated and in case of cooperation,

for example, in truck driving or CNC manufacturing, the roles

were clear: Human workers performed control and decision tasks,

machines and robots executed the mechanical tasks of production

and transportation. That situation however is changing as automa-

tion enters a new stage of AI applications (Wong et al. 2012; Musa

et al. 2014; Zhang et al. 2014; Knoll et al. 2016; Li et al. 2017;

Deng 2018). Robots, machines, and devices such as containers or

transportation equipment will be able to take informed and

advanced decisions without manual intervention, while the human

workforce takes a supervisory control and oversight role

(Castelfranchi and Falcone 2000; Cantor 2016; Crainic and Mon-

treuil 2016; Phan et al. 2017; Zhong et al. 2017). Consequently,

the qualification requirements for humans will migrate toward

cooperation with artificial intelligence applications within a

“know-when”-domain: Humans have to recognize and decide, for

example, when to override and stop automated applications in case

of potential danger or unforeseen changing conditions (Fischhoff

et al. 1978; Kim et al. 2011; Gurkaynak et al. 2016).

This development and the upcoming challenges embedded

therein are relevant for a large number of employees. For exam-

ple, in Germany, more than 2.9 million people are working in

the logistics sector, of which 868,000 in the land transport sector.

Although, for instance, automated truck driving technology is

available and the number of tests is rapidly increasing, human

drivers will still be needed for a long time. The further develop-

ment of HCI performance in the light of upcoming AI applica-

tions is a highly relevant topic (Koo et al. 2015; Weyer et al.

2015). How logistics in particular will be influenced by AI appli-

cations and HCI—considering aging and demographic challenges

in the transportation and logistics labor market—remains an

intriguing question (Nuzzolo and Comi 2014; Hasanefendic et al.

2015; K€

onigs and Gijselaers 2015).

Connecting these developments to sustainability and in particu-

lar to the triple bottom line approach (Schneider 2015; Brockhaus

et al. 2016), it is clear that social,environmental, and economic

sustainability are all affected by developments regarding AI appli-

cations in transportation and logistics. (1) So far, the social dimen-

sion has been largely neglected in research and practice as outlined

already by Seuring et al. (2008, p. 1545). Further contributions

include Ramos et al. (2014), Mani et al. (2016), and Sudarto et al.

(2017). Work conditions, security, and safety connected with AI

applications qualify as social dimension questions that require ade-

quate training (Missimer et al. 2017a,b; Sodhi 2015). (2) The eco-

logical dimension of sustainability is addressed as many AI

Corresponding author:

Matthias Klumpp, Fraunhofer Institute for Material Flow and

Logistics (IML), J.-v.-Fraunhofer-Str. 2-4, 44227 Dortmund, Ger-

many; E-mail: matthias.klumpp@iml.fraunhofer.de

Journal of Business Logistics, 2019, 40(3): 265–278 doi: 10.1111/jbl.12198

© 2019 Council of Supply Chain Management Professionals