UK Network Science Workshop

11 May 2018 (Fri) @ University of Bristol

Aim of the workshop

There are a significant mass of network science researchers in the UK. In this one-day workshop, we will discuss recent researches of the presenters (mostly from statistical physics communities) and discuss what we could do for expanding network science in the UK.

Programme up to date on 7 May 2018.


10:00-10:30 Registration --- Coffee/Tea

10:30-10:35 Opening (Naoki Masuda)

10:35-11:05 Lucas Lacasa (Queen Mary University of London)

"Detecting hidden layers in networks"

11:05-11:35 Guven Demirel (University of Essex)

"Identifying dynamical instabilities in supply networks using generalized modeling"

11:35-12:00 Coffee/Tea

12:00-12:30 Thilo Gross (University of Bristol)

"Stability of meta-foodwebs"

12:30-13:40 Lunch (Pugsley Lecture Theatre Foyer)

13:40-14:10 Tim Rogers (University of Bath)

"Heterogeneous micro-structure of percolation in networks"

14:10-14:30 Sadamori Kojaku (University of Bristol)

"Core-periphery structure requires something else in the network"

14:30-15:00 Vincenzo Nicosia (Queen Mary University of London)

"Probing network structure through random walks"

15:00-15:30 Coffee/Tea

15:30-17:00 Round-table discussion (led by Thilo Gross)

Registration (no fee): No need to register. Just come by. Lunch will be provided; I will guess the number of participants and order catering. Sorry if the amount of food is not enough!

Venue: Room 1.6, Queens Building, University of Bristol

Practical: Difficult to enter the building. The most accessible entrance faces University Walk, not Woodland Road or Cantock's Cl. See the map below.

Address: Woodland Road, Clifton, Bristol, BS8 1UB

How to travel to University of Bristol is found here

Organised by Naoki Masuda (University of Bristol, Department of Engineering Mathematics)

Funded by Collective Dynamics Research Group at University of Bristol

Talk Abstracts (optional)

Guven Demirel (University of Essex)

"Identifying Dynamical Instabilities in Supply Networks Using Generalized Modeling"

Supply networks are highly vulnerable to disturbances and disruptions, exposing the firms to instabilities and thus a high level of risk. Here, we apply Generalized Modeling (GM) to study the instabilities in inventory dynamics that develop due to the topology of the supply network. We first perform a bifurcation analysis to investigate the stability of some model network structures. We show that dyads and serial supply chains are immune to topology-induced instabilities. In contrast, in a simple triadic network, where a supplier acts as both a first and a second tier supplier, instabilities emerge from a wide range of bifurcations, i.e. saddle-node, Hopf and global homoclinic bifurcations. We then provide stability analyses of two extensive, contrasting real-world supply networks - a supply network of an industrial engine manufacturer and an industry-level supply network in the luxury goods sector. We find that suppliers with a high level of connectivity, e.g. assemblers and stockists, processing suppliers with bi-directional links, and suppliers that are located at the apex of cyclic network motifs, are identified as critical for stability and the spreading of disturbances in supply networks.

Lucas Lacasa (Queen Mary University of London)

"Detecting hidden layers in networks"

The architecture of many complex systems is well described by multiplex interaction networks,

and their dynamics is often the result of several intertwined processes taking place at different

levels. However only in a few cases can such multi-layered architecture be empirically observed,

as one usually only has experimental access to such structure from an aggregated projection. A

fundamental question is thus to determine whether the hidden underlying architecture of complex

systems is better modelled as a single interaction layer or results from the aggregation and interplay

of multiple layers. In this talk I will show that, by only using local information provided by a random

walker navigating the aggregated network, it is possible to decide in a robust way if the underlying

structure is a multiplex and, in the latter case, to determine the most probable number of layers. The

proposed methodology would also allow to select the optimal architecture capable of reproducing

non-Markovian dynamics taking place on networks, such as human or animal mobility. Applications of this

methodology in biophysics are finally discussed.