Physical Review E · 2008 · 14 citations · 13 references
BusinessDecision SciencesDecision MakingQueuing TheoryStochastic ProcessesBounded RationalityDynamic ProgrammingPoisson ProcessesQueueing ProcessesQueueing TheorySequential Decision-makingOne-dimensional DynamicsPriority AssignmentRational Agents
In order to explain the empirical evidence that the dynamics of human activity may not be well modeled by Poisson processes, a model based on queuing processes was built in the literature [A. L. Barabasi, Nature (London) 435, 207 (2005)]. The main assumption behind that model is that people execute their tasks based on a protocol that first executes the high priority item. In this context, the purpose of this paper is to analyze the validity of that hypothesis assuming that people are rational agents that make their decisions in order to minimize the cost of keeping nonexecuted tasks on the list. Therefore, we build and analytically solve a dynamic programming model with two priority types of tasks and show that the validity of this hypothesis depends strongly on the structure of the instantaneous costs that a person has to face if a given task is kept on the list for more than one period. Moreover, one interesting finding is that in one of the situations the protocol used to execute the tasks generates complex one-dimensional dynamics.
13
Wide area traffic: the failure of Poisson modeling
Vern Paxson, Sally Floyd · IEEE/ACM Transactions on Networking · 1995 · 3.7K citations · Full text
The origin of bursts and heavy tails in human dynamics
Albert-Ĺaszló Barabási · Nature · 2005 · 2.3K citations · Full text
Vladimir Rykov, M. Yu. Kitaev · International Journal of Stochastic Analysis · 1995 · 195 citations · Full text