Incident: Citi Bike Program Faces Major Software Failures and Operational Challenges

Published Date: 2013-06-11

Postmortem Analysis
Timeline 1. The software failure incident with New York City's bike-share program happened before June 11, 2013, as per the article published on that date. [19357]
System 1. Software from a company called 8D Technologies [19357] 2. Public Bike System Company's software [19357]
Responsible Organization 1. The software company 8D Technologies was responsible for causing the software failure incident in New York City's bike-share program [19357].
Impacted Organization 1. Riders of New York City's bike-share program [19357]
Software Causes 1. The software causes of the failure incident in the New York City bike-share program included technical errors in the system, such as docking stations being temperamental, refusing to accept bikes or process credit card information, and shutting down altogether [19357]. 2. The software from the company 8D Technologies used in New York's bike-share program was identified as a key issue, with other cities using the same operator but different software not experiencing similar widespread difficulties with docking [19357]. 3. The software problems were exacerbated by issues with the customer service phone line repeatedly failing, unreliable station information provided by an app, and delays in providing membership keys to annual subscribers [19357]. 4. The software malfunction also led to challenges in distinguishing between riders who struggled to return bikes and those who simply rode for too long, resulting in overage charges that needed to be rescinded [19357].
Non-software Causes 1. Overwhelming public demand leading to heavy use rates for bikes and stations, causing strain on the system [19357]. 2. Stations being both solar- and battery-powered, with their charges dependent on factors like weather and use rates for bikes, as well as the adjoining touch-screen device, leading to potential power drainage issues [19357]. 3. Inexperience of riders with the system contributing to operational troubles [19357].
Impacts 1. Many docking stations were temperamental, refusing to accept bikes or process credit card information, leading to inconvenience for users and potential revenue loss for the program [19357]. 2. Passers-by were able to pull a bike from a station without paying, indicating a security flaw in the system [19357]. 3. Riders had to test individual bike docks in search of one that works, leading to frustration and wasted time [19357]. 4. The customer service phone line repeatedly failed, making it difficult for users to report problems or seek assistance [19357]. 5. Station information provided by an app was unreliable, causing confusion among riders [19357]. 6. Many annual subscribers had not received their membership keys, impacting their ability to use the service smoothly [19357]. 7. The system had difficulty distinguishing between riders who struggled to return bikes and those who simply rode for too long, leading to overage charges for some users [19357].
Preventions 1. Conducting thorough testing and quality assurance of the software before the program launch to identify and address technical errors and glitches [19357]. 2. Implementing a more reliable and robust software system, possibly developed in-house or by a different software provider with a proven track record of successful bike-share programs in other cities [19357]. 3. Providing adequate training and support for users to navigate the system effectively, reducing user-related issues and errors [19357].
Fixes 1. Implementing more rigorous testing procedures for the software to identify and address technical errors before deployment [19357]. 2. Collaborating with experienced software development companies to improve the reliability and functionality of the system [19357]. 3. Enhancing customer service capabilities to address user issues promptly and effectively [19357]. 4. Conducting regular maintenance and updates to the software to ensure optimal performance and stability [19357].
References 1. New York City Transportation Department officials 2. Seth Solomonow, a spokesman for the department 3. Officials in Washington and Boston 4. Philip Pugliese, the bicycle coordinator in Chattanooga, Tenn. 5. Janette Sadik-Khan, the city’s transportation commissioner 6. Users of the bike-share program, including Abe Stanway, Erin Stabile, and others 7. Records from a lawsuit filed by Alta Bicycle Share against Public Bike System Company 8. Citi Bike Facebook page comments

Software Taxonomy of Faults

Category Option Rationale
Recurring one_organization, multiple_organization (a) The software failure incident having happened again at one_organization: The software failure incident occurred with New York City's bike-share program, specifically with the docking stations and the system's software. The program faced technical errors and problems with docking stations not accepting bikes, processing credit card information, shutting down, and even allowing passers-by to take bikes without paying. The software issues were severe enough that some users grew weary of the unreliable system and resorted to using private bikes instead [19357]. (b) The software failure incident having happened again at multiple_organization: The software failure incident in New York's bike-share program highlighted issues with the system's software provided by a company called 8D Technologies. Other cities like Washington and Boston, which also used the same operator (Alta Bicycle Share) but had software from a different company, did not experience similar widespread difficulties with docking or station failures. Additionally, Chattanooga, Tennessee, another city using the same software, also faced docking problems, albeit on a smaller scale due to the program's smaller size [19357].
Phase (Design/Operation) design, operation (a) The software failure incident in the New York City bike-share program can be attributed to design-related issues. The article mentions that the system had technical errors never experienced by bike-share programs in other major American cities, with many docking stations being temperamental, refusing to accept bikes or process credit card information, and shutting down altogether. The problems were linked to the software system's inability to handle public demand effectively, with issues such as passers-by being able to pull a bike without paying, riders struggling to find working docks, and long wait times for customer service [19357]. (b) Additionally, the software failure incident can also be linked to operational issues. The article highlights problems with the solar- and battery-powered stations, where the strength of their charges depended on factors like weather and use rates for bikes. The system faced challenges in distinguishing between riders who struggled to return bikes and those who rode for too long, leading to overage charges being rescinded. Furthermore, the customer service phone line repeatedly failed, station information provided by an app was unreliable, and many annual subscribers had not received their membership keys, indicating operational difficulties [19357].
Boundary (Internal/External) within_system (a) The software failure incident related to the New York City bike-share program was primarily within the system. The article mentions technical errors within the system, such as docking stations refusing to accept bikes or process credit card information, shutting down altogether, and passers-by being able to pull a bike without paying due to system issues [19357]. Additionally, the problems were attributed to the system itself rather than user inexperience. Officials in other cities using different software did not experience similar widespread difficulties with docking, indicating that the issues were specific to New York's system [19357].
Nature (Human/Non-human) non-human_actions, human_actions (a) The software failure incident in the New York City bike-share program was primarily attributed to non-human actions, such as technical errors and system malfunctions. The article mentions problems with docking stations refusing to accept bikes or process credit card information, shutting down altogether, and passers-by being able to pull a bike without paying due to the system issues [19357]. (b) However, human actions were also mentioned as a contributing factor to the software failure incident. The city's Transportation Department suggested that riders' inexperience with the system was contributing to their troubles, implying that user behavior played a role in the issues faced by the bike-share program [19357].
Dimension (Hardware/Software) hardware, software (a) The software failure incident occurring due to hardware: - The article mentions that the docking stations in New York City's bike-share program are both solar- and battery-powered, with their charges dependent on factors like weather and use rates for bikes, as well as the adjoining touch-screen device. Factors like weather and usage rates can impact the strength of the charges, affecting the functionality of the stations [19357]. (b) The software failure incident occurring due to software: - The article highlights various technical errors and problems plaguing the bike-share program's system, such as docking stations refusing to accept bikes or process credit card information, shutting down altogether, and passers-by being able to pull a bike without paying. Issues with the software have been a major factor in the system's unreliability and inability to handle public demand effectively [19357].
Objective (Malicious/Non-malicious) non-malicious (a) The articles do not provide any information indicating that the software failure incident related to the New York City bike-share program was malicious in nature. There is no mention of any intentional actions by individuals to harm the system [19357]. (b) The software failure incident related to the bike-share program in New York City appears to be non-malicious. The issues were attributed to technical errors, docking station problems, credit card processing issues, system breakdowns, and software unreliability. The problems were described as "inevitable kinks" by the Transportation Department, and the city officials suggested that riders' inexperience with the system contributed to the troubles. The failures were mainly due to system-related issues rather than intentional harm [19357].
Intent (Poor/Accidental Decisions) poor_decisions, accidental_decisions From the provided article, there is information related to both poor decisions and accidental decisions contributing to the software failure incident: (a) poor_decisions: The article mentions that the software failure incident in New York City's bike-share program was partly due to poor decisions made in the selection of software and operators. The program faced delays because "the software doesn’t work" as stated by Mayor Bloomberg [19357]. Additionally, the chosen operator, Alta Bicycle Share, had issues with the software provider, 8D Technologies, leading to docking problems and system failures in New York City's program [19357]. (b) accidental_decisions: The article also hints at accidental decisions or unintended consequences contributing to the software failure incident. The Transportation Department tried to downplay the issues as inevitable kinks and suggested that riders' inexperience with the system was a factor in the problems [19357]. This implies that there were unintended consequences or mistakes made in the implementation or user experience of the software system.
Capability (Incompetence/Accidental) development_incompetence (a) The software failure incident in the New York City bike-share program seems to be related to development incompetence. The article mentions that the system had technical errors of a magnitude never experienced by bike-share programs in other major American cities. Many docking stations were temperamental, refusing to accept bikes or process credit card information, shutting down altogether, and allowing passers-by to pull a bike without paying. The software issues were severe enough to lead to riders growing weary of testing individual bike docks, experiencing long wait times for customer service, and facing unreliable station information provided by an app [19357]. Additionally, the delays in the program launch were attributed to software issues, with Mayor Bloomberg stating that "the software doesn’t work" [19357]. (b) The software failure incident could also be considered accidental to some extent. The city's Transportation Department tried to minimize the issues by referring to them as inevitable kinks and suggesting that riders' inexperience with the system contributed to the troubles. However, a survey of programs in other cities indicated that New York's problems were more related to the system itself rather than user error. The article highlights that other cities using the same operator but different software did not face similar widespread difficulties with docking or station failures [19357].
Duration temporary The software failure incident related to the New York City bike-share program can be categorized as a temporary failure. The article mentions technical errors in the system, such as docking stations refusing to accept bikes or process credit card information, shutting down, and passers-by being able to pull a bike without paying [19357]. These issues are attributed to factors like the system's software and the inexperience of users, indicating that the failure was due to specific circumstances rather than being a permanent, inherent flaw in the software.
Behaviour crash, omission, value, other (a) crash: The software failure incident in the New York City bike-share program involved crashes where docking stations shut down altogether, refused to accept bikes or process credit card information, and had technical errors of a magnitude never experienced by bike-share programs in other major American cities [19357]. (b) omission: The software failure incident also included instances where passers-by were able to pull a bike from a station without paying, probably because the last user was unable to lock it back in place. Riders grew weary of testing individual bike docks in search of one that works, pedaling off to another station before the system eventually allowed them to end their trip [19357]. (c) timing: The software failure incident did not specifically mention timing issues where the system performed its intended functions too late or too early. (d) value: The software failure incident included failures where the system performed its intended functions incorrectly, such as not accepting bikes, not processing credit card information, and not allowing users to properly dock their bikes [19357]. (e) byzantine: The software failure incident did not exhibit behaviors of a byzantine failure where the system behaves erroneously with inconsistent responses and interactions. (f) other: The software failure incident also involved unreliable station information provided by an app, many annual subscribers not receiving their membership keys, and the customer service phone line repeatedly failing [19357].

IoT System Layer

Layer Option Rationale
Perception embedded_software The software failure incident related to the perception layer of the cyber physical system in the New York City bike-share program was primarily due to issues with the embedded software. The article mentions problems with the docking stations, such as them refusing to accept bikes, not processing credit card information, shutting down, and allowing passers-by to take bikes without paying. These issues point to failures in the embedded software controlling the docking stations [19357]. Additionally, the article highlights that the software issues were not attributed to user inexperience but rather to the system itself. Other cities using different software did not experience similar widespread problems with docking stations, indicating that the software used in New York's program was a significant factor in the failures [19357].
Communication unknown The software failure incident related to the New York City bike-share program does not explicitly mention whether the failure was specifically related to the communication layer of the cyber-physical system that failed. The articles primarily focus on technical errors with docking stations, credit card processing, bike locking mechanisms, and customer service issues. There is no direct mention of failures at the link level (physical layer) or connectivity level (network or transport layer) in the provided article [19357].
Application TRUE The software failure incident described in the article [19357] appears to be related to the application layer of the cyber physical system. The article mentions technical errors such as docking stations refusing to accept bikes or process credit card information, shutting down altogether, and passers-by being able to pull a bike without paying due to locking issues. Additionally, the article discusses problems with the customer service phone line, unreliable station information provided by an app, and delays in receiving membership keys for annual subscribers. These issues point towards failures introduced by bugs, operating system errors, unhandled exceptions, and incorrect usage, which are characteristic of application layer failures in a cyber physical system.

Other Details

Category Option Rationale
Consequence property, delay The consequence of the software failure incident described in the article is primarily related to delays (e). The software failures in New York City's bike-share program led to numerous issues such as docking stations not accepting bikes, problems processing credit card information, stations shutting down, users unable to lock bikes back in place, and riders having to test multiple docks to find one that works. These issues caused significant delays for riders, with some resorting to using private bikes instead of the bike-share system due to unreliability [19357].
Domain transportation (a) The failed system in the article was related to the transportation industry. The software failure incident was specifically about New York City's bike-share program, which experienced technical errors and problems with docking stations, credit card processing, and overall system reliability [19357].

Sources

Back to List