Authors:
Andreas Ahrens
1
;
Ojaras Purvinis
2
;
Detlef Hartleb
1
;
Jelena Zaščerinska
3
and
Diana Micevičienė
4
Affiliations:
1
Hochschule Wismar, University of Technology, Business and Design, Wismar and Germany
;
2
Kaunas University of Technology, Kaunas and Lithuania
;
3
Centre for Education and Innovation Research, Riga and Latvia
;
4
Panevėžys University of Applied Sciences, Kaunas and Lithuania
Keyword(s):
Buyers’ Burstiness, Independent Event, Gap Processes, Binary Customer Behaviour, Burstiness Estimation, Burstiness Measurement.
Related
Ontology
Subjects/Areas/Topics:
Aggregation, Classification and Tracking
;
Data Manipulation
;
Sensor Networks
;
Signal Processing
;
Statistical and Adaptive Signal Processing
Abstract:
Nowadays, bursty business processes are part of our everyday life. Bursty business processes include such processes
as selling and buying, too. One of the contemporary challenges business environment has to deal with
is monitoring and controlling of burstiness in business processes. Monitoring and controlling of burstiness in
business processes often leads to the optimization of business processes. Validation of the model for analysing
buyers’ burstiness in business processes revealed the need in optimisation of the proposed model, as the elaborated
model based on gap processes is complex for implementation, as well as for parameter estimation. For
optimization of the model for analysing buyers’ burstiness in business process, different levels of burstiness
in the process of buying are studied in this work. Different approaches to modelling buyers’ behaviours are
presented and evaluated in this work, too. The novel contribution of this work is based on the estimation of
burstiness. W
ith the proposed solution the level of burstiness can be estimated by taking the mean value and
the standard deviation of a gap sequence into account, which always exists for a given sequence. As a practical
application, the cash register of a medium size grocery shop in Lithuania is analysed. The novelty of this paper
is given by the comparison of different approaches to measuring burstiness in real process data. Directions of
further research are proposed.
(More)