Innovation Challenges

YCH Group Pte Ltd

YCH has cargo at the warehouse that we require to be scanned by the operators. Currently, in order to get all the information, the operator has to scan all the cargo individually. This current practice is both inefficient and time consuming. With this practice, human error is inevitable as the operator might miss out on scanning all the labels.

The existing solution in the market is using the camera to take a picture of all the labels on the cargo. The photo will then be used to perform a single transaction to extract the relevant information. However, this solution is not feasible if the number of labels exceed the frame of the camera.

A smart scanning system for the mobile device should have the ability to capture both the text and barcode on the cargo. In the case an inbound of 100 cartons, this system should enable the operator to get all the information of the 100 cartons with minimal scanning action(s). The system should be able to capture and recognise the relevant data. The data will then be converted into digital format and aggregate into an agreed output file (e.g. Microsoft Excel file). The Excel file should be accessible through the centralise database for easy reference.

What We Are Looking For

Our operators currently face the challenge of having to scan one carton label at a time which is time consuming. We are seeking a solution to improve the efficiency in scanning these carton labels.

The carton labels may have different kinds of layouts, and the proposed solution should be flexible enough to read a variety of layouts, such as a combination of text and barcodes on a label.

There may also be defective labels or those partially occluded by wrappings, which may pose a challenge for the scanning.

The proposed solution should be able to capture, recognise and convert the data into digital format and associate the respective fields, and aggregating the data into a Microsoft Excel file.

The proposed solution is to include the recommendation of the type of devices to be used for the scanning, eg. Android-based devices.

The working environment has normal lighting conditions and WiFi connectivity is available.

We are seeking the following levels of performance of the solution:

• Accuracy of 100% for barcode and QR Code, ~90% for text.
• 8 scans per second (single scans) / 1 scan per second (multiple barcodes and text)

A cloud-based solution is a preferred option as it allows the user to have the ability to access the documents from anywhere with internet connection.

Learn More

YCH Group Pte Ltd

YCH is seeking a technology to address the issue of inventory overflow. During peak periods, we may face the issue of insufficient storage space for these surges of incoming goods exceeding our warehouse capacity.

What We Are Looking For

We are seeking solutions that are able to mitigate the issue of inventory overflow.

We are open to various types of solution, for example:

• Sharing economy Airbnb-like platform to access a network for on-demand warehousing, fulfilment and logistics services
• Predictive Analytics to predict surges and recommended courses of action to mitigate inventory overflow
• Artificial Intelligence / Machine learning engines for storage infrastructure or algorithms to pre-emptively optimise space allocation of incoming goods

We are open to any other solutions that may address the issue of inventory overflow at our warehouses.

Learn More

Tristar Group

We are a company that owns ocean-going vessels such as tankers, gas carriers and dry bulk carriers. We are seeking an Internet of Things (IoT) platform to extract real time data from vessel’s equipment ashore and analyse the data for predictive diagnosis of the machinery problems on-board.

What We Are Looking For

We are seeking an IoT platform to extract vessel’s equipment data ashore for monitoring performance and diagnostics of machinery problems. The proposed solution should be able to predict a probable breakdown through trends and data analysis based on machinery readings and parameters.

Maintenance of the machinery will be carried out on-board; however, the analysis of the machinery performance will be done ashore which will determine the schedule of maintenance on-board.

Currently, the vessels undergo a planned maintenance schedule, with manual reporting of the vessel’s equipment performance performed by the vessel’s crew. Such a manual reporting of the machinery data introduces risk of human error and is therefore not ideal. Manual processes would also reduce the data frequency and the accuracy of the data recorded. This would hinder the probability forecasting of the machinery breakdown.

The solution should be able to extract data and determine the following:-

• Machinery failure due to the lack of adhering to the parameters and condition as recommended by the manufacturers (e.g temperature, pressure, amperage etc.)
• Machinery failure due to lubrication failure, or poor quality of lubrication oil
• Excessive vibrations leading to non-alignment, bearing failure, distortion etc.
• Machinery failure due to non-adherence to the planned maintenance as stipulated by the manufacturer
• Able to measure carbon footprint based on the vessels’ speed and consumption
• Monitoring fuel consumption and savings
• Tide / Weather predictions to assist in navigation
• Automatic vessel routing
• Database of all crew who have worked on the vessels & monitoring on-board and on-shore training of the crew
• Ship Inspection Report Programme (SIRE) and Port State Control (PSC) inspections
• Incident reports

For solutions that require machinery to be retrofitted with sensors, we shall get the machinery manufacturer to liaise with the solution provider to facilitate extraction of the data, and to advise on any restrictions or constraints.

With regard to wireless connectivity, our vessels have Very Small Aperture Terminal (VSAT) satellite communications that can be used to transmit data to and from the Internet.

Our vessels have data from the machinery collected on-board in electronic format.

The proposed solution is to be used by ship owners and ship managers.

The solution may be developed in any location; however, it has to be installed on-board ocean-going vessels.

Learn More

Teckwah Value Chain Pte Ltd

Teckwah Value Chain supports global clients to source, manufacture, assemble, kit and pack their final products. The turnkey kitting and packing step in particular is a labour-intensive and manual process.

The company currently faces the following issues:

• Turnkey processes consist of multiple manual steps to cater to different customer needs which involve a wide variety of combinations of items for packing
• Each customer’s requirements vary due to different/unique product packaging designs, and there are over 100 combinations of items for packaging
• Customers also have varying order patterns due to urgency and periodic surges (i.e. product launches)

What We Are Looking For

The company is seeking technologies/solutions that can address the following objectives:

• Identify work step process of manual activities for process improvement to achieve labour productivity and output quality consistency
• Modularity of the proposed solution to cater for different product configuration

Learn More

Tech-Link Storage Engineering Pte Ltd

As our Installation team is gradually allowed to commence work post Covid-19, it is crucial to re-imagine how we carry out work at customer’s site.

The goal is to ensure all daily tasks are catalogued and captured in addition to contact tracing and safe-entry guidelines put in place by the government.

The challenge is to have the ability to have a single platform, i.e. mobile device app, web-based software & etc. that combines Covid-19 and safety requirements.

What We Are Looking For

To develop a user-friendly app with the ability to capture the following:

1. To perform all safe entry requirement under a single platform/ app. (All personnel)
   • Scanning of Safe-Entry QR code
   • Temperature
   • Health declaration
   • Contact tracing
   • Social distancing measurement
2. Report of daily briefings & completed tasks (Supervisor level or above)
3. Inclusion of risk assessments write-up (Supervisor level or above)
4. Report of incident at site (Supervisor level or above)
5. All information must be digitalised to facilitate record keeping and retrieval.

Learn More

Sumitomo Corporation Asia & Oceania Pte Ltd

PT Monotaro Indonesia (Monotaro) is is an B2B ecommerce company, Joint Venture company of Monotaro Co.Ltd and Sumitomo Corporation Group.

On receiving data on goods from our suppliers, we need to categorize these items and map them to our existing product categories. This process is currently a manual one. There are numerous instances of incorrect categorization due to human-error and the process is time-consuming. We are therefore seeking a solution to automate the product categorisation of received goods from our suppliers.

We are interested in a software solution which can be deployed easily and provide a significant reduction in labelling errors, thus increasing productivity and enhance our overall work efficiency.

What We Are Looking For

The proposed solution should be able to accept incoming data from different suppliers, and then proceed to label the products as belonging to 1 of the 12 main categories. This proposed AI sorter system will automate the manual mapping process of assigning a product category label to an incoming product.

The incoming data from different suppliers could be in text such as the title and description, and image. The proposed solution must be able to recognise text and image data, to support the AI in identification of the most appropriate product category to the product. Descriptive text for the items could be either in Bahasa Indonesia or English.

In terms of training data for such AI system, we have internal data from our eCommerce platform. The smallest product category contains about 30k to 50k product records.

The proposed solution should incorporate data aggregation, cleansing and analytics.

Learn More

Sumitomo Corporation Asia & Oceania Pte Ltd

PT Monotaro Indonesia (Monotaro) is is an B2B ecommerce company, Joint Venture company of Monotaro Co.Ltd and Sumitomo Corporation Group.

To provide consistency and competitive product pricing to users of our eCommerce platform, we are seeking a solution which enables automated price optimization.

There is a lack of automated price optimising solution specific to the Indonesia’s online and offline markets. Here are some of the challenges:

• Some product categories are not standardised with unique product identifiers or displayed online.
• Some products may not have barcodes as they are not mandatory in Indonesia.
• Identifying products are a challenge in Indonesia as the names, descriptions and images may differ across the same products.
• In terms of offline markets, many of products in Indonesia do not have barcode or unique product identifiers that could be tied to price and other attributes.

What We Are Looking For

We are seeking a technology that may work through a combination of backend crawler of online and offline markets which is integrated to a system, where acquired product pricing can be matched to similar products. An output for the proposed optimized pricing for matched products can then be produced and automatically updated on our eCommerce platform.

Learn More

How can we develop a data analytics model that will facilitate optimisation and consolidation of inbound and outbound shipments in order to enhance cargo flow and make Singapore an ideal air cargo transhipment hub?

Learn More

How can we create a solution that allows real-time track and trace of air cargo movement in the warehouse?

Learn More

How can we develop an automated system that accounts for real-time inventory tracking of cargo containers and NMEs to provide real time tracking?

Learn More

How can we digitalise the cargo handling process to allow for monitoring and real-time tracking of shipments while integrating it with existing IT systems? 

Learn More

Pacific International Lines Pte Ltd

PIL Maintenance and Repair (MNR) department is responsible for the maintenance and repair of containers.

This includes vetting all repair recommendations submitted to PIL.

These repair recommendations are done by third party quantity surveyors who inspect PIL’s containers upon their arrival at any port.

If the surveyor discovers any damage that he deems as requiring a repair, he submits a repair recommendation to PIL with below

• Container number
• List of damaged components
• List of recommend repair actions
• Images taken of the container

Problem #1

Using the images uploaded for each recommendation, a PIL staff has to verify that:

• The container number tallies
• Each damaged component listed can be found in the images
• The recommended repair for each damaged component is reasonable given the extent of the damage detailed by the images

The images uploaded are not labelled to suggest which damage component it captures, and the PIL staff has to manually comb through all the images submitted for each repair recommendations to manually verify all the above pointers.

Problem #2

The next step in the container repair process after the container has been verified to require the specific repair job is to decide whether to repair the container locally or to reposition it overseas for a cheaper or higher quality repair.

The factors affecting the decision include:

• Type of the container needing repair
• Shipping schedule
• Availability of the type of repair and the difference in cost of repair in the different ports.
• Agreement rates of terminal cost of the port.

The team either sacrifices a huge amount of time to read through the details of the tariffs or to decide based on past experiences, which may not always provide the best course of action for the particular repair case.

What We Are Looking For

We are seeking a solution that automates and improves the productivity of the maintenance and repair damage assessment and repair verification process. The proposed solution should provide the following features:

• Automated recognition, labelling of survey results (images, reports etc.)
• Automated classification of damage
• Automated recommendation for type of repair
• Automated smart suggestion regarding repair location
• Automated smart decision on optimized repositioning of repaired equipment

Note that the images of existing damage components of the containers will be available to the solution provider for training.

Learn More

Pacific International Lines Pte Ltd

With bunkers being the highest OPEX item for any ship-owner/operator, optimizing bunker consumption and procurement could bring about substantial cost savings for the company.

This task is performed by shore-based vessel operators, who would collate and analyse various vessel performance indicators (e.g. vessel position, average speed, sea conditions, distance travelled, propeller slip, estimated times to destination, bunker remaining-on-board (ROB)) that are provided by vessel masters daily through a document known as the Noon Report. Based on historical performance data and/or that of similar class of vessels (i.e. sister vessels), a bunker optimization plan would be generated and communicated to vessel masters.

However, analysing large quantities of vessel performance data and deriving an optimisation plan takes considerable time. By the time the bunker optimization plans are sent to vessel masters, the plans might have been outdated as the vessel might have sailed off to a different location or the operating environment would have changed.

What We Are Looking For

Generating and Communicating Bunker Utilisation Plans

An application that receives and processes vessel performance statistics from the entire fleet of vessels, combines with SMHI weather statistics*, and prepares an optimised bunker utilisation plan for the vessel based on its historical performance data and that of its sister vessels.

The optimisation rules engine should be able and could run tirelessly 24 x 7 to provide timely optimization plans and suggestions to vessel masters for their proactive actions, removing the need for vessel masters to wait for advice from on-shore technical and operations team and giving significant cost savings from sub-optimal bunker consumption. 

Bunker Procurement and Inventory Management

The optimisation rules engine should also take into account of price movements in the global bunker market as well as the bunker inventory as it generates the bunker utilisation plan for vessel operators. From there, vessel operators would make an assessment on the advice before a bunker procurement decision is made.

Learn More

Lumens Auto Pte Ltd

Lumens Auto is a car leasing company specialising in long term leasing to drivers who are driving for ride hailing platform. Other than long term rental, we also do short term and corporate leasing. As the vehicles are used for ride hailing purposes, the drivers clock an average of more than 100,000 km/yr. Lumens Auto encounter the challenge of drivers encountering multiple accidents due to the long time spent on the road, in some instances, 5 accidents a year by a single driver. Apart from the risk of injury and even potential loss of life for the driver and passengers, this issue incurs revenue loss through expensive car repairs, along with reputational damage to the company. Insurance premium will also be increased due to the high accident rates. We are therefore seeking a cost-effective solution that will help to reduce the accident rate among drivers.

What We Are Looking For

Lumens Auto currently collects GPS data through as GPS device installed in all our vehicles. The GPS service provider is able to provide us with a lot of data, including distance travelled, harsh braking, etc. With these data, we are able to gain useful insights through data analytics.

We also have stand-alone cameras installed in the cars to collect video recording for use as evidence in the event of accidents. As the video recording are stored on an SD card, the data is automatically overwritten after 3-4 hours.

We have however not yet been able to leverage these data for analytics that provides insights or predictions that will help in reducing accident rates of our drivers.

For this challenge, we are open to testbed other devices such as On-Board Diagnostics (OBD) devices that may be installed in our vehicles to collect relevant data for analytics that will enable reduction of accident rates among our drivers.

We are open to any other solutions apart from those involving gathering data for analytics, eg. behavioural change motivation.

We have a fleet of mainly Japanese and Korean-make hybrid sedans and Multi-Purpose Vehicles (MPVs). The average duration of a driver’s shift is about 9-10 hours/day, with some drivers who can drive up to 14 hours a day. Most of our vehicles are rented out on a long-term basis, but we have a fleet of 10-20 vehicles that we keep idle as accident replacement cars which we can use for a POC project.

There are no geographical restrictions as to where the development is to be carried out, however installation has to be done in Singapore.

Learn More

GeTS Asia Pte Ltd

GeTS support the digital logistics space for Small & Medium Enterprises (SMEs) and Freight Forwarders (FFs). We aim to cover the purchasing and financing segment of the supply chain, and automate this segment with blockchain.

Why blockchain? Blockchain may be viewed as a secondary source of funds. It is decentralized, transparent, and it has smart contracts to enable auto-disbursements. Global trade financing is a wide and massive market, however it is highly dependent on human intervention and is not yet technologically advanced. It is not uncommon to find a long wait of customers often encountering rejection or no-response. It is largely a traditional industry, held back by slow and manual processes.

What We Are Looking For

We currently have a cloud-based machine-learning AI financing platform CALISTA, that enables automated financing, payment, and remittance. It handles ‘Know Your Customer’ (KYC) and audit checks and assessment of company credit applications, for fund disbursement. This includes a blockchain-enabled web and mobile application on Android and iOS, for trade financing. This application uses blockchain to prove ownership in the pledging of assets as collateral. The platform is implementing IMDA’s TradeTrust which is a framework that facilitates the interoperability of electronic trade documents exchanged between different digital ecosystems.

We are seeking collaborators to improve upon our current Blockchain solution,

The proposed technology should be compatible with IMDA’s TradeTrust framework and technology.

Learn More

GeTS Asia Pte Ltd

Getting the contact network, calling the carriers, freight forwarders via phone line and waiting on hold can be incredibly aggravating, not to mention time consuming. And for carriers, freight forwarders, phone centers are sometimes not the most efficient (and cost-conscious) way to address customer’s concerns.

On top of that, embracing the “new normal”, businesses have moved into communicating via instant messaging more and more. So far what we see in the market is that each of the carriers has social media channels for their business only. There is no one common social media channel that can allow different parties to discover each other and take the communication privately on and off the platform.

What We Are Looking For

A real-time communications and direct instant messaging chat-based solution is therefore sought, to enable efficient and seamless communications by our users. The solution should be integrated with our existing supply chain platform CALISTA.

Learn More

Continental Automotive Singapore Pte Ltd

There is a high demand internally to collect large amounts of video and image data, which is currently manually carried out, to be used for AI algorithm development. There is a lack of suitable and sufficient datasets from an automotive environment, both inside and outside the car.

Current Existing Method

Collection of video recording data in vehicle is carried out manually.

The manual collection data has certain restriction and gaps to be addressed, e.g.

• Inability to cover dangerous driving scenario on the road.
• Inability to cover all possible cases and combinations of driving behaviour and activities.
• Lack of flexibility (different camera position, field of view, colour/Infrared (IR) camera types, etc)

What We Are Looking For

We are seeking a technology or solution that is able to automatically generate realistic image and video data to be used in Computer Vision and Machine Learning Artificial Intelligence (AI) algorithm development.

Learn More

• Shipping details are printed on existing pre-printed form e.g. BL or manifest.
  
• Various customs departments require shipping details to be printed on port specific customs manifest format e.g. format specific to Myanmar, Philippines, India, and Kolkata.
  
• To cater to these numerous formats, shipping companies need development time & effort for each specific customs manifest format.
  
• It is time consuming and cumbersome to test the programs developed every time a new format is to be introduced/on-boarded.
  
• The effort for this is significant due to involvement of multiple parties in the integration process – e.g. developer for program development, users for testing, customs personnel for liaison on the requirements and also for integration testing.
  
• Each port has a specific format for the customs manifest. To develop N number of programs to meet each port specific requirement, the time & effort is N times.
  
• We are seeking a software solution that can translate the existing information populated in the BL or manifest to any specific customs manifest format without program development, which will save time and effort and reduce onboarding time significantly.

Singapore Post Limited

For over 160 years, Singapore Post (SingPost) as the country's postal service provider, has been delivering trusted and reliable services to homes and businesses in Singapore. Today, SingPost is pioneering and leading in eCommerce logistics as well as providing innovative mail and logistics solutions in Singapore and around the world, with operations in 19 markets. Building on its trusted communications through domestic and international postal services, SingPost is taking the lead in end-to-end integrated and digital mail solutions.

Almost 80% of postal addresses in Singapore are served by letterbox nests. The accuracy of the final delivery leg is dependent on the individual postman or postwoman. The fulfillment for ordinary (basic) mail is not traceable because there is no scanning activity.  A key performance indicator set by the regulatory authority requires 98% of successful postal deliveries to be done within a day of the item being posted (D+1).

To meet the myriad challenges and ensure that the D+1 requirement is met, SingPost launched a SMART Letterbox project in September 2019, that seeks to upgrade the letterbox infrastructure. The solution envisaged will eliminate the sortation activity within the letterbox nest, and generate a timestamp when delivery is completed. There are at least 12,000 buildings and hundreds of private residential blocks in Singapore.

It will take years to complete the implementation of the SMART Letterboxes nationwide, and deployment constraints have already been identified.

What We Are Looking For

We are seeking a solution that can implemented within the next 1 year to complement the SMART Letterbox project, by retrofitting existing letterbox infrastructure and/or introducing wearable technology. Examples of locations where this solution can be implemented include those with space constraints and/or constraints that relate to deployment, utilisation and lack of commercial viability. The proposed solution should eliminate human error in letter delivery, to ensure that the correct letter is deposited in the correct letterbox.

Learn More

How might we implement a chatbot equipped with AI and machine learning capabilities to handle common system issues on a 24/7 basis to enhance the user experience and response time?

Learn More

Nursultan Nazarbayev Foundation

Mining companies in Kazakhstan consists of a large number of employees. The workers must be provided on time the necessary equipment and special clothing. On average, 25 items are provided per employee, with the number of employees reaching 1,500. All the items must be monitored and accounted for, and the collected data must be converted into a report. One of these reports is the company's required order for special clothing from the contractor. The number of active companies from the extractive sector in Kazakhstan is 500.

What We Are Looking For

We are seeking an automated system for monitoring and accounting of special clothing for workers.

Learn More