Incident: Robotic Parking Garage Software Failure in South Florida, 2015

Published Date: 2015-11-28

Postmortem Analysis
Timeline 1. The software failure incident at the automated robotic parking garage in Miami Beach happened around five years before the article was published in November 2015 [53155]. Therefore, the estimated timeline for the software failure incident would be around November 2010.
System 1. Boomerang Systems' robotic parking garage system at Collins Avenue, Miami Beach [53155] 2. Park Plus robotic parking system [53155]
Responsible Organization 1. Boomerang Systems [53155] 2. Park Plus [53155]
Impacted Organization 1. Customers who parked their cars in the automated robotic parking garage [53155] 2. Crescent Heights, the property developer, which sued two manufacturers over the botched garage [53155] 3. Tenants at BrickellHouse who had to pay $28 a day to park elsewhere after the failure of the $16 million robotic garage [53155] 4. Boomerang Systems, the company that built the unsuccessful South Florida garages and later declared bankruptcy [53155] 5. Park Plus, the company hired to make fixes to the failed Collins Avenue garage [53155]
Software Causes 1. The software causes of the failure incident in the automated robotic parking garage included errors in the computerized system that parks and delivers cars, leading to malfunctions, cars being smashed, faulty machinery falling several stories to the ground, vehicles getting stuck for long periods, and the system bringing back the wrong cars [53155].
Non-software Causes 1. Lack of proper testing and readiness before opening the garage to the public, leading to malfunctions, smashed cars, and faulty machinery falling several stories [53155]. 2. Errors in design and construction of the automated parking garages, including issues with ramps, lifts, and shelves, as well as differences in design that were not properly understood [53155]. 3. Unfamiliarity of drivers with the technology, leading to problems such as wrong cars being retrieved, vehicles getting trapped, and delays in the parking process [53155]. 4. Overcrowding and high demand for parking spaces, especially during peak times, causing bottlenecks and delays in retrieving cars [53155]. 5. Issues with the physical infrastructure of the garage, such as cars dangling off platforms, being squashed in shafts, and the inability to handle the traffic volume efficiently [53155].
Impacts 1. Cars were smashed, and faulty machinery fell several stories to the ground, leading to significant property damage and safety hazards [53155]. 2. Vehicles were stuck for extended periods, forcing garage operators to pay for customers' taxis, resulting in financial losses [53155]. 3. The garage was not ready to be open to the public, causing inconvenience to customers and tarnishing the reputation of the property developer [53155]. 4. The garage could only handle 16 cars per hour instead of the intended 60, leading to operational inefficiencies and customer dissatisfaction [53155]. 5. The software failure incident resulted in the bankruptcy of the company that built the garages, Boomerang Systems, and subsequent liquidation of its assets [53155]. 6. The incident led to pending litigation between the building developer, Harvey Hernandez, and Boomerang Services, impacting their business operations and financial stability [53155].
Preventions 1. Thorough testing and quality assurance procedures before deploying the automated parking system could have helped prevent the software failure incident [53155]. 2. Conducting pilot tests with a smaller number of cars to identify and address any potential issues before scaling up the system for full operation [53155]. 3. Providing proper training and education for users to ensure they understand how to interact with the automated parking system correctly, reducing the chances of user-generated delays and errors [53155]. 4. Collaborating closely with experienced automated parking system developers who have a successful track record in implementing similar projects to avoid common pitfalls and mistakes [53155].
Fixes 1. Implementing simpler technology designs like the one used by Auto Parkit, which has been successful in the automated parking industry [53155]. 2. Ensuring proper user education and training to prevent driver-induced delays and system jams [53155]. 3. Conducting thorough testing and quality assurance to address software and hardware mishaps before deployment [53155].
References 1. Crescent Heights - the property developer [53155] 2. Boomerang Systems - the company that built the unsuccessful South Florida garages [53155] 3. Park Plus - another robotic parking company involved in fixing the garage [53155] 4. Harvey Hernandez - the building developer involved in pending litigation with Boomerang Services [53155] 5. Ryan Astrup - the director of Park Plus [53155] 6. Ron Lowy - a Miami lawyer who sued two parking companies on behalf of Crescent Heights [53155] 7. Christopher Alan - from Auto Parkit, a company with several automated parking garages [53155] 8. Gil Dezer - a South Florida developer with plans for an automated garage in the Porsche Design Tower [53155] 9. Casey Jones - a former chairman of the International Parking Institute [53155]

Software Taxonomy of Faults

Category Option Rationale
Recurring one_organization, multiple_organization (a) The software failure incident having happened again at one_organization: - The company Boomerang Systems, which built the two unsuccessful South Florida garages, declared bankruptcy and announced voluntary liquidation of its assets [53155]. - Boomerang Systems had seven robotic parking projects, and one of the garages built for 139 cars on Collins Avenue sat unused for five years due to software and hardware issues [53155]. (b) The software failure incident having happened again at multiple_organization: - The article mentions other attempts at self-parking garages around the country facing embarrassing software and hardware mishaps [53155]. - Instances of failures in automated parking garages were reported in different locations such as New Jersey and Maryland, indicating a broader issue with the technology implementation [53155].
Phase (Design/Operation) design, operation (a) The software failure incident related to the design phase can be seen in the article where it mentions that the automated parking garages in South Florida faced spectacular debacles due to errors in design and attempts to duplicate foreign successes without understanding the differences in design that can make or break a project [53155]. The article highlights that some garages designed to park and deliver cars quickly ended up bringing back the wrong cars, trapping vehicles, taking too long, and even damaging automobiles due to design flaws in the automated systems. (b) The software failure incident related to the operation phase is evident in the article where it describes how drivers, unaccustomed to the technology, contributed to errors in the operation of the automated parking garages. For example, drivers walking away without pushing a button to tell the garage to park their car caused jams in the system for everyone else, leading to delays and inefficiencies in the operation of the automated parking systems [53155].
Boundary (Internal/External) within_system (a) The software failure incident described in the articles is primarily within_system. The failure of the automated robotic parking garages in South Florida was attributed to malfunctions, faulty machinery, errors in design, and software and hardware mishaps within the system itself. The incidents involved cars being smashed, machinery falling several stories, vehicles getting stuck, wrong cars being retrieved, and long delays in returning cars to customers. These issues were a result of problems with the automated parking technology and the inability of the system to handle the traffic efficiently [53155].
Nature (Human/Non-human) non-human_actions, human_actions (a) The software failure incident occurring due to non-human actions: The articles describe a software failure incident related to automated robotic parking garages where malfunctions led to cars being smashed, faulty machinery falling several stories to the ground, vehicles getting stuck for long periods, and the garage not being ready for public use [53155]. These failures were attributed to errors in the automated system itself, such as the garage not being able to handle the traffic efficiently, cars being returned in longer times than expected, and the system getting jammed when users did not follow the correct procedures [53155]. (b) The software failure incident occurring due to human actions: The articles also mention human actions contributing to the software failure incident in the context of drivers not being accustomed to the technology, which led to delays and errors in the automated parking process [53155]. Additionally, some garage builders tried to replicate foreign successes without understanding the differences in design that could impact the project's success, leading to embarrassing software and hardware mishaps [53155].
Dimension (Hardware/Software) hardware, software (a) The software failure incident occurring due to hardware: - The article mentions incidents where faulty machinery in the automated robotic parking garage fell several stories to the ground, leading to malfunctions and cars being smashed [53155]. - In Hoboken, N.J., a Cadillac plunged six stories, and a Jeep dropped four stories due to hardware failures in a robotic garage [53155]. (b) The software failure incident occurring due to software: - The article discusses how some automated parking garages in the U.S. faced embarrassing software and hardware mishaps, leading to issues like bringing back the wrong cars, trapping vehicles, and damaging automobiles [53155]. - It is mentioned that the garage on Collins Avenue in South Florida faced software-related problems, with test runs showing that the garage takes about seven minutes to retrieve cars instead of the intended three minutes, and it can handle only 16 cars per hour instead of the planned 60 cars [53155].
Objective (Malicious/Non-malicious) non-malicious (a) The articles do not mention any malicious intent behind the software failure incident related to the robotic parking garages in South Florida. The failures were primarily attributed to technical issues, design flaws, and operational challenges rather than intentional harm to the system [53155]. (b) The software failure incidents related to the robotic parking garages in South Florida were categorized as non-malicious. The failures were a result of errors in design, hardware malfunctions, software glitches, and operational inefficiencies rather than any deliberate attempt to harm the system [53155].
Intent (Poor/Accidental Decisions) poor_decisions, accidental_decisions From the provided articles, the software failure incident related to the automated robotic parking garage in Miami Beach can be attributed to both poor decisions and accidental decisions. 1. Poor Decisions: The incident highlights poor decisions made by garage builders who tried to duplicate foreign successes without understanding the differences in design that can make or break a project. This lack of understanding led to errors and mishaps in the automated parking system [53155]. The company that built the unsuccessful garages, Boomerang Systems, declared bankruptcy and voluntarily liquidated its assets, indicating financial mismanagement and poor decision-making [53155]. 2. Accidental Decisions: The article mentions that errors were common in the United States due to drivers being unaccustomed to the technology, indicating unintentional mistakes in implementing the automated parking system [53155]. The delays and malfunctions in the robotic garage resulted in cars being smashed, machinery falling several stories, and vehicles getting stuck, suggesting unintended consequences of the system's failures [53155].
Capability (Incompetence/Accidental) development_incompetence, accidental (a) The software failure incident in the articles can be attributed to development incompetence. The automated robotic parking garage in Miami Beach experienced malfunctions, cars being smashed, faulty machinery falling several stories, vehicles getting stuck, and the garage not being ready for public use [53155]. The issues were a result of errors in the design and implementation of the automated parking system, indicating a lack of professional competence in ensuring the system's reliability and safety. (b) Additionally, the failure can also be considered accidental as there were instances of cars being damaged, falling off platforms, and getting trapped due to software and hardware mishaps [53155]. These incidents were not intentional but occurred as a result of unexpected failures in the automated parking system.
Duration temporary The software failure incident related to the automated robotic parking garage in Miami Beach was more of a temporary nature. The malfunction lasted for hours, leading to cars being smashed, faulty machinery falling several stories to the ground, and vehicles getting stuck for extended periods, requiring garage operators to pay for customers' taxis [53155]. Additionally, the garage experienced delays, closures, and operational issues, such as bringing back the wrong cars, trapping vehicles, and damaging automobiles [53155]. These issues indicate a temporary software failure incident caused by specific circumstances rather than a permanent failure introduced by all circumstances.
Behaviour crash, omission, timing, value, other (a) crash: The software failure incident described in the articles can be categorized as a crash. The automated robotic parking garage experienced malfunctions that led to cars being smashed, faulty machinery falling several stories to the ground, and vehicles getting stuck for extended periods [53155]. (b) omission: The software failure incident can also be categorized as an omission. The garage failed to park cars correctly, leading to delays, cars being trapped, and the wrong cars being retrieved, omitting the intended function of efficient and accurate parking [53155]. (c) timing: The timing of the software failure incident can be considered a factor as well. The system was not able to deliver cars within the expected timeframe of three minutes, taking much longer, which caused inconvenience to users and created bottlenecks in the process [53155]. (d) value: The software failure incident can be attributed to a failure in value as well. The system did not perform its intended function of delivering cars correctly, resulting in damaged automobiles, cars dangling off platforms, and squashed in shafts, indicating incorrect performance [53155]. (e) byzantine: The software failure incident does not align with a byzantine failure, which involves inconsistent responses and interactions. The incident described in the articles primarily focuses on malfunctions, delays, and incorrect parking rather than erratic or inconsistent behavior [53155]. (f) other: The other behavior exhibited by the software failure incident is a failure in reliability and safety. The incident led to cars being smashed, machinery falling, vehicles getting stuck, and the need for manual intervention to park cars, highlighting a lack of reliability and safety in the system [53155].

IoT System Layer

Layer Option Rationale
Perception sensor, actuator, embedded_software (a) Sensor: The software failure incident related to the robotic parking garage in Miami Beach involved malfunctions where cars were smashed, faulty machinery fell several stories to the ground, and vehicles were stuck for hours, leading to garage operators having to pay for customers' taxis. These issues indicate failures related to the sensors used in the automated parking system [53155]. (b) Actuator: The article mentions instances where the automated parking system brought back the wrong cars, trapped vehicles, took a long time to return cars, and even damaged automobiles. These issues point towards failures related to the actuators responsible for moving and retrieving the cars within the garage [53155]. (c) Processing_unit: The software failure incident described in the articles does not explicitly mention failures directly related to the processing unit of the automated parking system. Therefore, there is no specific information provided regarding failures introduced by processing errors. (d) Network_communication: The articles do not highlight any failures specifically attributed to network communication errors in the context of the robotic parking garage incident in Miami Beach. Therefore, there is no relevant information provided regarding failures introduced by network communication errors. (e) Embedded_software: The software failure incident involving the robotic parking garage in Miami Beach showcases issues such as cars being stuck, wrong cars being retrieved, delays in returning vehicles, and overall inefficiencies in the system. These issues are indicative of failures related to the embedded software controlling the operations of the automated parking system [53155].
Communication unknown The software failure incident described in the articles does not specifically mention whether the failure was related to the communication layer of the cyber-physical system that failed. The focus of the incident was on the malfunctioning and operational issues of the automated robotic parking garage, such as cars being smashed, faulty machinery falling, vehicles getting stuck, wrong cars being retrieved, and delays in car retrieval times. The articles primarily highlight mechanical problems, hardware mishaps, and operational challenges rather than explicitly discussing failures at the communication layer of the system.
Application FALSE The software failure incident described in the articles does not specifically mention that the failure was related to the application layer of the cyber physical system. The focus of the incident is on the mechanical and operational failures of the robotic parking garages, such as cars being smashed, machinery falling, vehicles getting stuck, wrong cars being retrieved, and delays in car retrieval. Therefore, it is unknown whether the failure was related to the application layer based on the information provided in the articles.

Other Details

Category Option Rationale
Consequence property, delay The consequence of the software failure incident described in the articles includes: (d) property: The software failure incident resulted in property damage, specifically to cars. The malfunctioning robotic parking garage caused cars to be smashed, damaged, and trapped, leading to financial losses for both customers and operators [53155]. (e) delay: The software failure incident caused significant delays for customers trying to retrieve their cars from the malfunctioning robotic parking garage. Customers experienced long wait times, with some having to wait 45 minutes to an hour to get their cars, well beyond the expected retrieval time of three minutes [53155].
Domain transportation, construction (a) The failed system was related to the transportation industry, specifically automated parking garages designed to revolutionize parking systems for vehicles [53155]. The incident involved malfunctions in the robotic parking garage, leading to cars being smashed, faulty machinery falling several stories to the ground, vehicles getting stuck for extended periods, and customers experiencing significant delays and inconvenience [53155]. (m) The failed system could also be categorized under the construction industry, as it involved the development and implementation of cutting-edge parking projects within luxury residential high-rises and commercial buildings [53155].

Sources

Back to List