Vehicle Interior Detection (1)

▤ 목차

Proof of Concept

반응형

728x90

The Intention of the Project

Picture from https://www.socarcorp.kr/about

The Socar app facilitates a structured and user-centric workflow that simplifies the car rental process into three main stages: renting, accessing, and returning the vehicle. The first stage, renting a car, begins with clients selecting a date, time, and location that aligns with their travel needs. This step provides flexibility to users, allowing them to plan their schedules with precision. Once the details are entered, the app utilizes its database to display all available vehicle options, enabling clients to choose the type of car that best suits their preferences, whether it be an economical sedan for daily commutes or a spacious SUV for group travel. Furthermore, Socar enhances customer convenience by offering two options for retrieving the vehicle: clients can either visit a designated Socar Zone, strategically located for easy accessibility or request delivery of the vehicle to a specific location such as their home or workplace. This dual approach ensures that the service caters to a wide range of users, from those who prefer self-pickup to those who value doorstep convenience.

 

This first step in the process demonstrates Socar’s commitment to providing a seamless digital experience that empowers users to make informed choices without needing physical assistance or visits to a traditional rental office. By integrating these features, the app reflects Socar’s broader strategy of leveraging technology to streamline the rental process while maintaining a high level of personalization and flexibility for its diverse customer base.

 

The second stage, accessing the car, is designed to ensure that clients can independently verify the vehicle's condition before beginning their journey. Upon arriving at the designated location or receiving the vehicle, users are required to perform an initial inspection of the car’s exterior. This involves taking photographs of the vehicle’s exterior from multiple angles and uploading them to the app. These images serve as both a record of the car’s condition prior to use and a safeguard against potential disputes regarding damage claims during the rental period. By implementing this step, Socar not only protects its assets but also fosters transparency and trust with its customers.

 

Following the exterior inspection, clients are prompted to assess the vehicle's interior and operational condition. This includes checking critical elements such as the vehicle dashboard for warning lights, ensuring the proper functioning of safety features like seat belts, and confirming that the overall cleanliness and usability of the car meet acceptable standards. These actions empower users to identify and report any pre-existing issues directly through the app, enabling prompt resolution by Socar’s support team if necessary. This self-service approach aligns with Socar’s goal of minimizing human intervention while ensuring that vehicles remain in optimal condition for each rental.

 

This phase underscores the company’s emphasis on user responsibility and transparency. By requiring users to actively engage in verifying the vehicle’s condition, Socar not only reduces the operational burden on its personnel but also encourages clients to take greater accountability for maintaining the car’s state throughout the rental period. In turn, this contributes to a more efficient and reliable car-sharing ecosystem.

 

The final stage of the process, returning the car, is designed to be as efficient and user-friendly as the previous stages, ensuring that the vehicle is returned in good condition and that the rental process is completed seamlessly. Upon arrival at the designated parking lot or return location, clients are expected to park the vehicle and photograph its exterior once again. This step is crucial for documenting the car's condition at the time of return, providing a record that ensures any potential damage or discrepancies can be identified and resolved swiftly. By having users take these photos themselves, Socar avoids the need for physical inspections by staff, thereby streamlining the return process and reducing costs associated with labor.

 

Once the exterior is documented, clients are required to finalize the return through the app. This involves confirming that the vehicle has been returned on time, is parked in the correct location, and meets the company's cleanliness standards. The app then processes the return, and clients are prompted to complete any final steps, such as confirming the time of return and verifying that no personal items have been left behind. In the case of a flawless return, the transaction is completed with minimal interaction, providing a smooth and hassle-free conclusion to the rental experience.

 

This system is particularly advantageous for Socar, as it reduces the need for physical checks by customer service representatives, enhancing operational efficiency and lowering overhead costs. Additionally, by automating the return process through the app, Socar eliminates the possibility of human error or inconsistencies in the vehicle check-in process, ensuring that clients are not penalized for issues that may have arisen during the rental period. The self-service nature of this step is aligned with the company's commitment to maintaining a frictionless user experience, further reinforcing Socar’s reputation as a leader in the car-sharing industry.

Limitations of the Service

While the app’s design is intended to provide a smooth and efficient user experience, the reality of its implementation presents several practical challenges. Socar’s system operates under the assumption that all clients will follow the outlined processes and return the vehicles in a clean and operational state. In an ideal world, every customer would return the vehicle exactly as they received it, ensuring that the vehicle is spotless and free of any faults. This would create a flawless experience for all parties involved, with minimal need for intervention from customer service personnel. However, this is often not the case in practice. Human nature and varying levels of responsibility among users mean that some clients may neglect to clean the interior or report vehicle issues before returning the car. As a result, when a new user accesses the vehicle, they may encounter a car that is dirty or has unresolved mechanical issues, leading to a negative initial experience with the service.

 

These situations often lead to customer service inquiries, where affected users contact Socar to report issues with the vehicle. In most cases, the affected clients are directed to another available car. While this helps resolve the immediate problem, it does not fully address the root cause. Clients who experience such disruptions may become dissatisfied with the service, and their perception of Socar may be negatively impacted. Even though the company enforces penalties for those who fail to maintain the cleanliness of the vehicles, the new client’s first experience with a dirty or malfunctioning vehicle can leave a lasting impression. This negative experience, if repeated, can lead to a gradual decrease in app usage and customer retention.

 

Moreover, these recurring issues also contribute to a higher operational cost for Socar. The company must continually manage customer complaints, redirect users to alternative vehicles, and process penalties for unclean returns. These manual interventions, while necessary, add complexity to the app's otherwise streamlined process, and may not be sufficient to address the underlying issues effectively. This scenario also highlights a gap in the existing system: there is no automated mechanism for ensuring that vehicles are inspected for cleanliness and functionality before they are rented out again. This gap represents a clear opportunity to enhance the user experience and improve the overall service by addressing these challenges through technology.

 

To address the issues caused by unclean or faulty vehicle returns, Socar implemented a penalty system designed to hold clients accountable for maintaining the vehicles' cleanliness and condition. Under this system, users who fail to return vehicles in a satisfactory state are charged additional fees, such as cleaning or repair costs. This approach aims to deter clients from neglecting their responsibilities while reinforcing a sense of accountability among the platform's user base. However, while the penalty system serves as a corrective measure, its effectiveness in preventing recurring issues remains limited.

 

From the perspective of a new client, encountering a vehicle in poor condition is a significant inconvenience that directly impacts their experience with the service. Although Socar’s customer service team often resolves such incidents by redirecting affected clients to another vehicle, the initial negative impression can be difficult to recover from. For many users, the inconvenience caused by the disruption outweighs the remedial measures provided. Over time, repeated instances of such experiences may erode customer trust and loyalty, resulting in a gradual decline in app usage and overall retention rates.

 

Moreover, while the penalty system ensures that offenders are financially penalized, it does little to improve the experience for subsequent users who must deal with the consequences of another client’s actions. This revealed a critical gap in Socar’s existing process: the lack of a systematic and automated mechanism to verify the vehicle’s interior cleanliness and operational condition during the return process. Without this step, the company cannot guarantee that each vehicle meets its service standards before being assigned to the next client. This gap not only undermines the company’s commitment to providing a seamless user experience but also represents an untapped opportunity to enhance the service through innovative solutions.

Potential Resolutions

In light of the identified challenges, our team considered several potential solutions to enhance the vehicle return process and address the gap in interior cleanliness verification. One of the initial proposals was to expand the app’s functionality by requiring clients to capture detailed photographs of the vehicle’s interior during the return process. These photos, covering critical areas such as the seats, floor, and dashboard, would be submitted via the app for manual review by Socar’s customer service team. This method would allow the company to ensure a higher standard of cleanliness and operational condition before making the vehicle available to the next user.

 

However, despite its thoroughness, this approach was not without drawbacks. A manual verification process would necessitate additional personnel to review the uploaded images, significantly increasing operational costs. Moreover, the additional step of taking and submitting interior photos could extend the return process, potentially creating delays and diminishing the convenience that Socar customers value. For a company that markets itself as a seamless and automated car-sharing platform, the introduction of a labor-intensive manual process risked conflicting with its core value proposition of efficiency and ease of use.

 

Additionally, the reliance on human oversight in the verification process introduced the possibility of delays and inconsistencies. Given the volume of vehicles in circulation at any given time, ensuring that each interior inspection was conducted promptly and uniformly would require a substantial investment in human resources and training. These challenges highlighted the need for an alternative solution that could maintain the integrity of the return process without sacrificing efficiency or profitability.

 

Recognizing the limitations of manual verification processes, the project team turned its attention to technology-driven solutions that could streamline the vehicle return process without compromising on quality or efficiency. One proposed alternative involved implementing an automated warning system within the app during the return stage. This system would display a reminder or notification, warning users that penalties would be imposed if the vehicle was returned in an unsatisfactory condition. While this solution required minimal additional investment and was easy to integrate into the existing platform, its effectiveness relied heavily on user compliance. There was still a significant risk that clients who were indifferent to penalties or unaware of the standards would fail to meet the cleanliness requirements.

 

Another concept involved allowing users to designate a returned vehicle for “maintenance review” if they were unable or unwilling to ensure the vehicle's readiness for the next client. This option would temporarily mark the car as unavailable for rental until it had undergone inspection and cleaning by Socar staff. Although this approach addressed the problem of unclean vehicles being handed over to new users, it introduced the risk of misuse. Clients could intentionally or unintentionally mark vehicles for maintenance, leading to unnecessary downtime and a reduction in fleet availability, ultimately impacting revenue.

 

Both options underscored the inherent difficulty of relying solely on user-driven or manual solutions to ensure quality standards. These methods, while feasible in theory, posed risks to Socar’s operational efficiency and profitability. As a result, the need for a more robust and automated system became increasingly evident—one that could seamlessly inspect and verify vehicle cleanliness and functionality without relying on human intervention or user discretion.

Problem Objective

After evaluating various potential solutions, the team determined that leveraging artificial intelligence (AI) was essential to address the challenges of inspecting vehicle interiors efficiently. Unlike manual processes, which could significantly increase operational costs and delay the vehicle return process, an AI-driven approach offers the advantage of automated image analysis, ensuring that vehicles meet cleanliness and functionality standards promptly. This decision aligns with Socar’s overarching goal of providing a seamless and user-friendly car-sharing experience while maintaining high operational efficiency.

 

The primary objective of the project is to integrate an AI model into the existing app workflow. This model would analyze photos of the vehicle's interior, submitted by users during the return process, to determine whether the car requires maintenance. By doing so, the app can quickly identify vehicles that are not rental-ready and flag them for further action, ensuring a smooth and positive experience for subsequent customers. This automation eliminates the need for employees to manually review images, thereby minimizing return processing time and enhancing the overall user experience.

 

From the company’s perspective, the integration of an AI-driven system also provides operational benefits. By categorizing maintenance needs into levels of severity—such as light maintenance (e.g., cleaning minor debris) and heavy maintenance (e.g., a full car wash or addressing mechanical issues)—Socar can optimize resource allocation. For instance, light maintenance can be addressed on-site by personnel, while heavy maintenance tasks, such as addressing faulty vehicles, can be delegated to Car Care, Socar’s subsidiary specializing in vehicle servicing. This approach not only ensures efficient issue resolution but also reinforces Socar’s ecosystem by keeping vehicle-related services in-house. By preventing reliance on outsourced maintenance, the company can maintain better quality control, reduce costs, and strengthen its brand identity.