Pages

16 December 2019

Embrace the App Store: Leveraging Smartphones for Army Maintenance

Andrew Shaughnessy

Against the backdrop of a generation of Soldiers who grew up tethered to smartphones, the Army remains rigidly analog in many of its systems. While there are numerous potential examples of how smartphone applications could make Army systems more accessible, few seem as immediately viable for disruption as unit maintenance. A dedicated Army smartphone maintenance app, guiding operators through weekly maintenance inspections, would alleviate many of the challenges encountered while trying to diagnose and resolve equipment faults. With the current emphasis on readiness, there is no reason that the Army should tolerate the existing shortcomings of our analog maintenance processes.

Few institutions in the Army are as universal or protected as the venerated tradition of “Motor Pool Mondays.” Across the force, Monday mornings are a protected slot of time where units hold formations, maybe receive some words of encouragement from their leadership, and then Soldiers immediately descend on their equipment to conduct Preventative Maintenance, Checks, and Services (PMCS). Soldiers take their printed Technical Manuals (TM) and go through a prescribed series of checks to ensure their equipment, primarily vehicles, will perform optimally. Soldiers will top off any vehicle fluids, test that electrical systems function properly, and annotate any vehicle faults on printed equipment inspection worksheets (DA Form 5988-E). Empowered by the Army’s centralized logistics software and database, Global Combat Support System-Army (GCSS-A) these 5988s are printed that morning and accurately reflect historical issues for each vehicle. After thoroughly inspecting their respective pieces of equipment, these analog 5988s are collected, reviewed, and then turned in to the Battalion’s Forward Support Company (FSC) to be digitalized in GCSS-A. From here, faults that can be fixed with parts on hand are remedied throughout the week or, in the absence of the right bench stock, have the correct parts placed on order. Leveraging cutting edge innovations in supply chain management, it is only a matter of time until parts arrive and equipment is brought up to a Fully Mission Capable (FMC) status.


While that system sounds simple enough in theory, the reality often falls far short of these expectations. Soldiers routinely lack printed TMs, skip necessary PMCS steps, and turn in illegible 5988s missing necessary data. Many of the physical TMs Soldiers use during Motor Pool Mondays are outdated, weathered and torn becoming barely readable, and are often few and far between. While the natural rebuttal to this dilemma is to simply order more, the TM ordering process has proven to be slow and often unresponsive. Units will often print off exclusively the PMCs chapters of the TM, but this excludes troubleshooting information and often only survives a few weeks before needing to be reprinted for that Soldier. With the PMCS chapter for a standard HMMWV being fifty-one pages long, printing one for every vehicle operator proves to be a heavy burden on organic office supplies. Many Soldiers will try to use PDF versions of the TM on their smartphones, but TM file sizes are often prohibitively large and smartphone PDF readers are not optimized for scrolling through them. As a result, a lot of weekly PMCS is sloppily done without stringently following the steps from the TM.

Even if heavily supervised, it is not uncommon to catch operators running vehicles for a few minutes and then copying over the same faults from the previous week. While Motor Pool Mondays also include supervisors such as NCOs, for units with a large number of vehicles, there simply is not enough leader bandwidth to ensure every vehicle receives the same quality of PMCS. Furthermore, even if the PMCS steps are followed closely, the TMs prescribe only basic diagnosis and not troubleshooting in that chapter of the TM. Unless an operator is determined to explore other portions of the TM or has a personal knowledge about vehicles, such faults get annotated on 5988s in only very general terms. For example, operators regularly describe vehicles that will not start as having inoperable batteries, rather than trying to isolate a cause against the myriad of electrical issues that could explain this.

Low quality PMCS translates into low quality 5988s, which in turn prevents maintenance teams from best allocating their time or ordering the correct repair parts. Many 5988s that get turned in are illegible, use incorrect nomenclature, are missing item numbers, or are physically damaged by the elements during PMCS. FSCs then have to contend with hundreds of pages of varying quality paperwork as they push to digitalize faults in GCCS-A. This in turn, often stretches their bandwidth to the extreme, so faults get missed and never get digitalized in GCSS-A. In turn, parts are not ordered against them, and vehicle operational statuses suffer as the same faults get annotated in unreadable handwriting, week after week. Collectively, these are not new problems. A Rand study from 1981 describes the same challenges in nearly the same terms then as they exist describes the same challenges in nearly the same terms then as they exist today. While many units may contend that their maintenance program performs better than described above, a casual survey of Soldiers reveals that cynicism over maintenance is the norm rather than the exception.

In the age of smartphones, there is no reason this system has to be this way. The Army should create a PMCS smartphone app, available on both iOS and Android, which walks operators through the PMCS steps. This is not a new idea. The Army aspired to make an app for conducting generator PMCS back in 2013. Looking through the various smartphone app stores, it does not look like this ever fully materialized. U.S. Army Combined Arms Support Command (CASCOM) produced a PMCS app, but it is quite literally just a few short animated videos on how to fill out a 5988. Some commercial PMCS apps exist, but at face value they appear to be low quality, featuring a narrow range of vehicles, and are paid apps. There are several PMCS apps on the GoCanvas platform, purportedly created by the Army, but video demonstrations of these apps appear clunky and outdated. In general, the idea of a PMCS app looks like something that has been routinely toyed with, but never pursued seriously.

Since the overwhelming majority of Soldiers possess personal smartphones, a PMCS app would be a widely accessible tool to improve the quality of unit maintenance. The Army already makes a suite of smartphone apps so this would not be uncharted territory. Furthermore, most FSCs already employ ruggedized tablets and laptops as Maintenance Support Devices (MSD) to store digital maintenance TMs. Many of the Army’s apps are admittedly clunky, so prioritizing an intuitive user interface would be essential to ensuring it actually gets used. One potential model for a user interface would be to mirror the right-swipe left-swipe dynamic found on apps like Tinder. With so many apps featuring swiping prominently in their user experience design, this is an empirically validated interface that many Soldiers would find highly intuitive. Soldiers could select a vehicle type from a drop-down menu, then the app would walk them through the correct PMCS steps for that vehicle. A Soldier would select their type of vehicle, conduct the PMCS steps by swiping right on FMC items, left where faults need to annotated. If a fault is present, the app would let the Soldier select a specific problem listed from a drop-down menu and describe more specific issues in a comments box. For more complex faults like electrical issues, it could guide operators through basic troubleshooting procedures to either resolve the issue or at least refine the specific problem at hand. At the conclusion of the PMCS, the app could generator a template of what a perfect 5988 for that vehicle would like. If the Army desires to keep this as a stovepipe system, the operator could then copy this perfect 5988 template over to a physical 5988 to turn in. With a perfect 5988 example present, Soldiers could take their time transcribing accurate information to the physical 5988, reducing many of the present pitfalls. Should the Army be comfortable integrating it with other systems, these 5988s could either be sent digitally to the Executive Officer and FSC representatives, or even eventually integrate with GCSS-A. By facilitating higher quality 5988s with greater specificity on faults, many of the worst burdens on FSCs ranging from interpreting handwriting to intuiting vague faults would be alleviated. This in turn, would facilitate efficiency in how maintenance teams allocate time for troubleshooting and repairs, as well as reducing the number of faults that get missed as 5988s become digitalized. If eventually integrated with GCSS-A, the burden of digitalizing faults could become largely automated.

Besides supplementing issues with physical TMs, a smartphone app would also add useful functionality. PMCS steps could eventually feature high resolution photos of the checks being conducted at hand, visual inspection guides for items like tire serviceability, or even feature how-to videos. Another useful feature would be a time stamp because an operator can simply copy over faults if not being closely supervised during PMCS, a procedural app serves as a forcing function to go through all of the steps in order. If the length of time from initiating the steps to concluding the PMCS is timed, it provides a quick litmus test as to how thorough a PMCS was. If an NCO supervisor checked a timestamp for a vehicles PMCS and saw that the entire PMCS lasted under five minutes, they would immediately know that the operator was improperly conducting the PMCS. The combination of forcing adherence to the proper PMCS steps in sequence, as well as monitoring the actual length of time taken, would make it vastly easier for supervisors to enforce quality PMCS.

The Army’s preventative maintenance magazine, PS Magazine, has already transitioned to being published in the form of a smartphone app. Unfortunately the app does not have a strong user interface and because it has to be downloaded by itself, is not likely to be read by many Soldiers. If there were a dedicated PMCS app intended to be used weekly, PS Magazine could be integrated as a secondary feature. If a Solder just completed a PMCS on a M119A3 howitzer, the app could then display relevant content related to optimal M119A3 maintenance. By integrating PS Magazine illustrations directly with the equipment undergoing PMCS at the time, the odds of operators reading and leveraging it would go up significantly.

Another major benefit of a PMCS smartphone app would be the amount of raw data that could be collected and analyzed. As long as the Army was willing to provide some level of server support, historical data from each PMCS could be accurately recorded and cataloged. If vehicle performance were also tracked, such as an Executive Officer annotating which vehicles broke down during a field exercise, it could be analyzed against PMCS data for insight. The Army already invests in predictive maintenance analytics that leverage machine learning for vehicles like the Bradley Fighting Vehicle. In the case of the Bradley, sensors on the vehicle itself are used to collect data to train algorithms. Conceptualizing it broadly, a Soldier performing a PMCS from a checklist is the ultimate multi-source sensor collecting serviceability data. Whereas the predictive analytics for the Bradley is limited to the types of data being collected by its sensors, storing all PMCS data would provide a broad perspective not captured by electronic sensors. If fault data were recorded for every single PMCS over time, and compared against performance data, machine learning algorithms could potentially be trained to identify which minor faults would correlate with eventual major faults under certain conditions. While IBM already has conducted proofs of concept with the Stryker demonstrating the utility of predictive maintenance and the Logistics Support Activity (LOGSA), a new database of PMCS data pulled from the app would provide the perfect launching point to expand horizons. Whereas GCSS-A does not seem to track historical fault data to individual vehicles or performance outcomes, a database of PMCS data could be tailored from its origins to train machine learning algorithms. A PMCS smartphone app, backed up by servers storing historical PMCS data for each vehicle, could provide an excellent data set to advance the burgeoning field of predictive maintenance analytics in the Army. Empowered by this insight, Army logisticians could make better informed decisions about when to conduct scheduled vehicle services, what parts to forecast ordering, and what types of maintenance behaviors genuinely correlate with optimal outcomes.

The United States Air Force has already demonstrated some proof on concept on integrating personal devices, under the banner of Bring Your Own Device (BYOD), in supporting aircraft maintenance. Creating a smartphone maintenance app, designed to replace physical Technical Manual’s PMCS chapters, could offer significant benefits to the Army’s unit maintenance programs. It would resolve many of the existing issues from conducting PMCS with the physical TM, translate into higher quality 5988s, and be designed to force strict adherence to proper PMCS procedures. It would not entirely replace physical TMs, but it would resolve many of a FSC’s enduring pain points in garrison where most maintenance happens. Should this PMCS data integrate with other Army digital maintenance systems, or have its own database support, it would be well suited to support predictive analytics. While Motor Pool Mondays are, and should remain, an immutable professional tradition, there is no reason they need to be executed using outdated tools. Embracing the ubiquity of smartphones and the expertise of a generation raised on app stores is perfect way to modernize this battle rhythm. With many Soldiers spending their weekends swiping left or right, there is no reason they can’t do that on Monday mornings as well.

No comments:

Post a Comment