Author Profile

Finding a Partner in a Disaster

Jun Zhuang, Industrial and Systems Engineering, SUNY at Buffalo

Summary

To illustrate the application of the proposed model in a practical environment, in the remainder of this section we will discuss the model from the perspective of an emergency manager assisting in Joplin, Missouri. Over 100,000 volunteers have poured in from around the country since an EF-5 tornado hit Joplin, MO in 2011 and left over 7000 structures damaged or destroyed. This presented a challenge to organizations involved in disaster relief and volunteer management as they struggled to ensure that the groups of volunteers that claimed they would help actually arrived and assisted in the effort. This is a common issue during the recovery phase when many new buildings are being built and a large number of volunteers are pouring into the impacted area.

Because construction projects are large, and typically require many volunteers over the project life, agencies that manage such projects often work with multiple groups (both large and small) to bring in the necessary number of volunteers. However, this is a fragile issue for both parties because incompatibility can lead to a loss of time and resource. For this set of examples, we look at a game between an NGO managing construction projects, and an outside group offering volunteers; each agency is attempting to maximize their impact in the disaster area with a constrained set of resources (projects and volunteers). Once two agencies have found each other, they have to decide whether or not to actually work together. For NGOs involved in reconstruction, a key component of the strategy is to have a consistent and reliable work-force. Skilled and unskilled volunteers are the primary labor force for this work. Volunteers can come from a variety of agencies including businesses, faith-based organizations/networks, government agencies, and other community organizations. This is an important issue since the number of volunteers can easily run into the tens or hundreds of thousands.

Individuals with expertise/responsibilities in the following areas have helped create the case study:
  1. Players in Group A (1/2 of the class)
  2. Players in Group B (1/2 of the class)
  3. Game Manager (course instructor)

Key teaching points:
  1. How does the size of an agency impact the likelihood of deception in the partnership?
  2. What was the cut-off point for reaching out to the player with known compatibility?
  3. Where players with large agencies more likely to be hesitant to offer a partnership?
  4. What players were the quickest to make two partners?

How this example is used in the classroom:
  1. Post information about partnerships
    • Names of the Groups: Explain that group members are only compatible with each other
    • Number of tokens: Each player is given some number of tokens to indicate agency size. Ideally agency sizes should range from 2*r to 2¬*(r + number of students).
    • Cost of Outreach: There is a cost to trying to find a partner. It costs r tokens to ask another player to form a partnership. This amount is paid to the game manager
    • Benefit of partnership: When two players agree on a partnership they go to game manager to make the partnership "official" and have an impact on the community. The impact on the community is calculated as the average of the two agency's current size
    • Cost of incompatibility: any partnership across the groups will reduce the impact of the partnership by c points
    • Any tokens not invested in a partnership at the end of the game will be returned to the game manager without any positive benefit for the player.
  2. Hand out slips of paper to students with the following information. This paper is private and the information cannot be shared with anyone
    • Which group they are (hence compatibility grouping)
    • The group of two other students (one the same, one different)
  3. Hand out another piece of paper with the size of the agency in tokens. This is public information
  4. Have students find two partners. They must spend a token to attempt a partnership, independent of the outcome. They can say whatever they want about compatibility as long as they do not show the secret slip to anyone besides the game manager.
  5. When a partnership is accepted, both players show their agency cards to the game manager who then states the impact of the partnership
  6. This cycle repeats until all players have two partners or have 0 tokens remaining
  7. At the end of the game, scores are calculated per agent as the agency impact divided by the initial agency size

References

D. McEntire. Issues in disaster relief: progress, perpetual problems and prospective solutions. Disaster Prevention and Management, 8(5):351-361, 1999.

J. Coles. Modeling and Simulating the Network Behavior of Agencies during Disaster Relief Operations. Buffalo, NY: University at Buffalo, 2014.
J. Coles, J. Yates, and J. Zhuang. Case study in disaster relief: A descriptive analysis of agency partnerships in the aftermath of the January 12th, 2010 Haitian earthquake. 46:67-77, 2012.

J. Coles and J. Zhuang. Decisions in disaster recovery operations: A game theoretic perspective on organization cooperation. Journal of Homeland Security and Emergency Management, 8(1): Article 35, 2011.

Joplin Public Information Office. Fact sheet - City of Joplin: Joplin Missouri hit by ef-5 tornado on May 22, 2011. Online. Accessed on March 21, 2014, February 2013. URL http://www.joplinmo.org/DocumentCenter/View/1985.

M Hager and J. Brudney. Volunteer management practices and retention of volunteers. Online report. http://www.urban.org/publications/411005.html, accessed on march 21, 2014, The Urban Institute, 2100 M St NW, Washington, DC, June 2004.

R Cnaan and T Cascio. Performance and commitment: Issues in management of volunteers in human service organizations. Journal of Social Service Research, 24(3-4):1-37, 1998.

Supporting Files