Job Seekers - Collaborate or not?

1 collaborator

Comments and Questions

Description

## WHAT IS IT? The purpose of the model is to study how collaborative and non-collaborative search strategies affect success of finding a job offer good enough to accept it. ## HOW IT WORKS ### Background Agents live in a torus-shaped space and wander around randomly in search for potential jobs. Available jobs are represented with green-colored patches. Patches can be arranged randomly or in a clustered manner. ### Agents Agents have their resource called "energy", and use this resource at each step. When an agent runs out of energy, it gets discouraged and stops searching, changing its color. When an agent steps on a green patch, it submits an application and acquires energy, and if energy is larger than a certain threshold, the agent is hired, stops moving and changes its color and the color of the patch (the job is not available anymore). ### Decision Agents can form networks by randomly dropping links to 10% of the total number of agents and collaborate within the network. Collaboration means that when an agent is hired, it invites other agents to come to the neighborhood of the job patch. ### Evaluation The decision about the best strategy is made based on the number of people who find jobs, the number of people who get discouraged, and the total sum of applications agents submitted. ## HOW TO USE IT ### Main controls Collaborate switch controls the choice of strategies. The observer can select the number of job-seekers. The observer can select the number of jobs available, and the number of clusters they're arranged in. For a totally random spacial distribution of jobs, the observer can select the maximum number of clusters (which is equal to the number of jobs). ### Plot The model displays a plot to keep track of the number of available jobs, the number of agents that get discouraged, the total number of applications submitted, and the number of agents that found a job. ### Other controls Agents have a random amount of energy between zero and the number chosen by the user with the Max-energy slider. When submitting an application to a job, an agent receives the additional amount of energy that the user can select with the Energy-from-application slider (this amount cannot exceed the Max-energy). Agents take steps of a random length, within the range that the observer can set with the max-distance-per-tick. ## THINGS TO NOTICE Calculate the difference of results (jobs found, discouraged agents, applications) depending on the chosen strategy. The number of jobs found should be equal to the number of patches that turn grey. ## THINGS TO TRY The observer is encouraged to try different strategies, as well as different cluster settings. The model will also bring different results by using the other sliders. ## EXTENDING THE MODEL Try to use one-directional links instead of bi-directional links. Develop a third strategy of competition, where instead of helping each other, agents can compete between each other. Design the model so that jobs are not a fixed resources, but can appear over time. ## NETLOGO FEATURES No unusual feature was used. ## RELATED MODELS Mushroom hunt, Wolf and sheep. ## CREDITS AND REFERENCES This model was created as a project as part of the ABM class at Data ScienceTech Institute. Created on December 12th, 2016 by Laurent Berder.

Posted almost 8 years ago

Click to Run Model

globals [ applications jobs_found discouraged remaining_jobs]
;OtherGlobals   Max-energy    Energy-from-job    N-job-searchers
;               N-jobs   distance-per-tick    Vision    collaborate
breed [job-searchers job-searcher]

job-searchers-own [attempts energy]

to setup ;what happens when you click "setup"
  clear-all
  ask patches [set pcolor 9]
  grow-jobs
  set-default-shape job-searchers "person" ;; create job-searchers and set them up
  set remaining_jobs N-jobs
  create-job-searchers N-job-searchers
  [
    set color violet
    setxy random-xcor random-ycor
    set energy random Max-energy
    set attempts 0
    if collaborate[create-links-with n-of (N-job-searchers / 10) other job-searchers] ; create links (both-way links) to 10% of job-searchers at random for collaborative strategy
  ]
  set applications 0
  reset-ticks
end 

to go ;what happens when you click "go"
  ask job-searchers [find-job]
;set measures
  set remaining_jobs count patches with [pcolor = green]
  set jobs_found count patches with [pcolor = 7]
  set discouraged count job-searchers with [color = red]
;end go
  if count job-searchers with [color = violet] = 0 [stop]
  ;if ticks = 400 [stop]
  tick
end 

to grow-jobs
  ;ask n-of N-jobs patches [set pcolor green] ; set jobs totally randomly
  ask n-of N-clusters patches [ask n-of (N-jobs / N-clusters) patches in-radius 5 [set pcolor green]] ; set jobs randomly in clusters
end 

to find-job
  if (energy < 1000) and (energy > 0) ; keep looking as long as they don't have a job
       [
         rt random-float 50 ;; continuous random right turn up to 50°
         lt random 50 ;; discrete random left turn up to 50°
         let turn-distance random max-distance-per-tick
         fd turn-distance
         set energy energy - 1 * turn-distance / 2
       ]

  if pcolor = green
  [
    let recruit random 10
    set energy energy + Energy-from-application
    if recruit > 8 [
      set pcolor 7
      set energy energy + 1000 ; when you get a job, you get a bonus energy
      set color blue
      set shape "face happy" ]
    set attempts attempts + 1
    set applications applications + 1

    if collaborate[
    let found-job patch-here
    ask link-neighbors with [color = violet] [move-to one-of patch-set found-job] ; if you find a job, tell your friends to come where you found it
    ]
  ]
  if energy < 1 [
            set color red
            set shape "face sad"] ; once their energy runs out, they get discouraged stop searching
end

There is only one version of this model, created almost 8 years ago by Laurent Berder.

Attached files

File	Type	Description	Last updated
Job Seekers - Collaborate or not?.png	preview	Preview for 'Job Seekers - Collaborate or not?'	almost 8 years ago, by Laurent Berder	Download

This model does not have any ancestors.

This model does not have any descendants.

NetLogo