Summaries - Office of Research & Innovation
Research Summaries
Back Optimally Locating MARFORRES Units
Fiscal Year | 2015 |
Division | Research & Sponsored Programs |
Department | Naval Research Program |
Investigator(s) | Salmeron-Medrano, Javier |
Sponsor | NPS Naval Research Program (Navy) |
Summary |
The co-PIs and their students will conduct research to improve the current MCRO model with (a) an enhanced mathematical model that better represents the MARFORRES Realignment problem, and (b) more accurate data. We next describe these items in detail: "a. Mathematical model development. We will integrate the following new specifications into the mathematical model: - Unit composition: The MCRO model assumes a unit's performance in a given area is solely based on the ratio of the size from all units in the area to its recruiting population. That is, ""unit size"" drives the decisions of which units to move, and where they are moved. However, attrition rates and other performance measures show that, currently, two collocated units might be performing very differently in the accomplishment of their missions. We will redefine unit's performance in MCRO by weighing in additional measures which may make an area to become more or less suitable for a unit. Specifically, we will incorporate unit requirements for specific rank, gender, prior vs. non-prior active duty composition, and military occupational specialty (MOS), in addition to the current measure for overall demographics. - Cost: The MCRO model assumes the user provides the areas to which each unit can be relocated. This user's input currently takes the form of ""yes"" or ""no."" In the first case, the optimization process can relocate the unit to the area based on demographic considerations, and in the second case it cannot. At no point in this decision there is consideration of the associated relocation cost, or the possibility that the unit can be moved if, for example, new facilities are built in the area. We propose to factor in our model costs for relocation (moving) expenses for some personnel and equipment, MOS retraining, facility upgrading, new military construction and operation of the facilities. - Time-phased relocation: MCRO's output is ""static,"" that is, it ignores when the units are actually moved over time. In reality, any relocation process is carried out during the span of up to ten years due to economic, operational and logistic constraints. We will investigate an extension to incorporate a time-phased relocation process, so MCRO could be used to optimize not only the relocated units but also the timing of those relocations. For example, our model could make provision to limit the number of moves to ""no more than five units per year,"" ""no more than 1,000 billets"" per year, ""no more than a certain cost per year,"" and the like. - Others specifications: We anticipate our discussion with MARFORRES planners will suggest additional factors to model, for example: - unit collocation or other dependency requirements - other special requirements (e.g., ""among three given units, select two to relocate and one to deactivate"") b. Data Collection and Analysis. An important difficulty in developing Brisker's MCRO model has been the lack of consolidated data. We will analyze existing and new data as follows: - Recruitable population: Our data from Marine Corps Recruiting Command (MCRC) is at the level of county boundary. However, this does not match MARFORRES' concept for an area: a 100-mile radius from a home training center (HTC) that the unit occupies. The BAH study simulates recruitable populations for each HTC area, but those populations overlap in many regions (in some cases up to 30 HTC areas covering the same spot). Brisker reconciles these data sets for MCRO using certain assumptions, but more work and/or validation on the topic is needed. - Area target ratio: MCRO currently uses a default value of 1-to-400 recruiting ratio as the target for all areas. This number is based on the MARFORRES' total billets and the 2011 recruitable population in the country. However, we anticipate that area-specific recruiting goals would greatly improve the model results. For example, there exist current units with low performance despite they are located in areas largely satisfying the 1-to-400 ratio, which indicates that either the ratio is inadequate for the area, or that other factors (such as those described in paragraph (a) above) are affecting the performance of the unit. We will work on defining means to calculate those ratios by area, and will compare the model results. - New data: The new specifications we propose to integrate (see (a) above) will require significant amounts of data. Based on the experience from the MCRO model development, we anticipate this as a significant challenging task. We will search for approximations based on existing information (e.g., new facility construction from past BRAC studies) when the actual data are not readily available." |
Keywords | |
Publications | Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal |
Data | Publications, theses (not shown) and data repositories will be added to the portal record when information is available in FAIRS and brought back to the portal |