When it comes to Government projects, few companies are generally able to receive orders directly. This is because government projects have strict requirements on manufacturers and they need to go through multiple layers of screening and testing. Only after passing the final examination can orders be given to manufacturers for production. In addition, the number of products in government projects can be very large, and the quality and delivery time of products is strictly controlled. In cases where there is a failure to deliver goods on time or other circumstances, it is necessary to be prepared to compensate for the corresponding amount of funds needed in the case of a breach of contract according to the terms of the contract signed, lest you end up blacklisted by a Government and be unable to participate in bidding for other projects in the future. Therefore, a strong factory is needed to complete such a task.
In early 2019 we began negotiations for an ODM learning machine project from the government of Turkmenistan. In order to promote Turkmenistan’s economic development and create job opportunities, this project had a specially purchased batch of learning machines to be given to the first-grade students. Therefore, we needed to prepare 80,000 sets of learning machine software and all of the assembly materials, with spared parts shipped in bulk to the country so they could assemble the finished products. The plan was for the goods to be delivered on time at ht end of June. In the later stage, special technicians needed to be equipped to provide technical guidance and support for the assembly of learning machines in the Turkmenistan factory.
In this non-commercial ODM order we encountered a number of thorny complications. As a children’s toy manufacturer with 15 years of experience in OEM&ODM as well as in independent development and R&D departments we were prepared to deal with these problems and successfully handled them. We believe that through sharing about the difficulties we handled it will illustrate our adaptability and timeliness in solving any problems, also providing a deeper understanding of the ODM children’s toy project.
I. Factory Inspection
Because this was a government project, it had strict requirements for all aspects of the factory. Our customers needed to check the factory with a third party. We needed to prepare the equipment and documents that were required in advance according to the customer’s test content. The contents that need to be prepared for this factory inspection are as follows:
2. Quantity of production equipment in the factory
|3. Internal test equipment project|
| 4. Passed the approved quality management system certificate,
factory certificate and product certificate
|5. Factory organization chart|
|6. Production process flow chart|
|7. Map of factory floor area|
Our factory covers an area of about 22,800 square meters, has 10 modern production lines, employs more than 500 employees, includes an aging test, drop test, SMT machine, and other product production and testing equipment. We deliver high-quality orders and always ensure our product quality and that we met customer needs.
1. Factory Plan
2. The Company’s Organizational Chart
3. Factory certificate
Our factory passed the inspection of ICTI, BSCI, GSV, SEDEX, ISO9001, SEDEX.
4. There is also the product certificate.
All our products are strictly in accordance with international environmental protection and safety requirements from materials to the end of the process and have passed the certification of the European Union, the United States, and other countries as well.
II. Order Requirements
After the order is confirmed, the customer provides a set of samples for us to make samples after completion of the 3D picture of the product. The detailed order requirements are as follows:
III. ODM software and hardware development of the learning machine
- Software Development
Our learning machine has 60 educational programs, with the language only in English. Hence, for our engineering department, to change this was not only a heavy workload but also a big technical test for our software development and design. In the process of manufacturing the learning board, the first problem we encountered was in the development of learning machine software.
Step 1: Organize and Translate Information
Our engineers needed to write the corresponding codes of the 60 educational programs in the English version into a table and provide them to the customer. We then let the customer fill in the corresponding Turkmen and Russian translations in the table. The corresponding code was used as the naming of the recording.
Step 2: Check the Translation and Recording Provided by the Customer
After the customer completed the translation and the recording was sent again, we needed to check whether the customer provided all the translation and recording needed in the form. If there was a missing translation or a lack of a recording, the customer needs to add it.
Step 3: Programming
After all the translation and recording materials were complete, the software engineer began to enter the programming according to the code one by one. This process took about 30 working days.
Step 4: Preliminary Testing and Burning Software
After completing the software programming and preliminary testing, the software engineer burned the software to the main control IC and gave it to our hardware engineer.
In the process of testing the software, we encountered the following problems:
(1) System halted. When the game switched languages, it caused the system to stop and sometimes crash.
(2) The picture was blurred. After switching between different languages, some programs appeared to be blurry or incomplete.
(3) Content confusion. There were some projects in Turkmenistan and Russian, and there were still some words and recordings that do not match.[Insert Picture Six]
It was a lot of work to correct these problems, but we were able to solve these issues and optimize the software.
The optimizations were as such:
There were some projects where the language translation and the recording content did not match correctly. The software engineer needed to enter the re-translated content into the system so as to find the corresponding code and to confirm whether the code was written correctly; if it was incorrect, the software engineer needed to rewrite the code and replace the previous wrong content. This process needed to be carefully examined, so that there were n more content replacement errors later which could cause chaos in the program.
For software crashes and blurred images engineers needed to conduct multiple screenings, check for software vulnerabilities, and constantly adjusted and optimized the software programs until the software vulnerabilities were resolved and the software could run normally. In the process of software development and design of the learning machine, our structural engineers first created 3D model drawings according to customer requirements. After the model was established, the structural engineer evaluated the product.
- Hardware Development
In the design process of running program software development, the learning machine model and circuit board design were also in progress. Our design department then designed the circuit board route of the learning machine according to the model (when designing the circuit board route, we needed to consider the current standard of Turkmenistan, the frequency standard of sound, etc.). After the roadmap was designed, engineers needed to assemble a circuit board for debugging, testing whether the current of the circuit board could withstand the operation of the software, whether the current frequency of the learning machine met the customer’s standard and so on. If the content was not up to standard, the engineer needed to continuously optimize and adjust settings until all the problems had been solved.
IV. Production of the Learning Machine
After the model of the product was confirmed, we arranged the shell proofing of the hand-board according to this model. After all the software design of ODM was optimized, we burned the program running the software into the master control IC, and then our hardware engineers welded other electronic components onto the circuit board, and then assembled the finished product. After many tests, we found some problems:
- Battery: When someone would unscrew the battery cover, the battery would bounce easily.
- Loudspeaker noise.
- The speed of keyboard response and picture translation was slow.
Because the design of the battery card slot was set according to the customer’s sample, our engineer changed the battery cover into the form of a buckle and fixed it with screws. The other two problems were finally optimized and solved by the engineer after several test modifications. The software ODM of the learning machine product was then completed, and the next step was the production process of the learning machine.
V. Mass Production of the Learning Machine
Although the goal of this project was the shipment of semi-finished products, the details of the production still needed to be paid close attention to, they were not any less important than a shipment of finished products. Through the cooperation of each department of our factory, this crucial link was successfully completed in the end. So how did we effectively control production efficiency and quality? Let’s cover that now.
- Time Control and Quality Inspection of Product Materials
We did not arrange the production of materials such as the screen of the learning machine, conductive glue, hardware, and other small accessories, and all were handed over directly to the procurement department for the purposes of procurement. The purchasing department looked for suppliers who often cooperated with each other according to the BOM List provided by the engineering department. Throughout this process, the procurement department needed to ensure that these materials could arrive within the specified time. The purchased materials then needed to be subjected to an IQC inspection by the quality department. Strictly according to the “Inspection Documents”, the inconsistent data needed to be evaluated according to the assembled and semi-finished products. IQC then would make a pass mark for the products that passed the inspection. If they were disqualified, the IQC then would submit these products to the quality supervisor to assess whether the defects that had occurred had an impact on the final quality of the product. Should an impact be noted, these materials were then handled according to different prescribed standards. Materials that would have had a serious impact on the product were then directly returned or repurchased. These materials that met requirements were to be exported to Turkmenistan for assembly so we needed to ensure that no problems would occur there, as a replacement would not be possible at that point. This necessitated much more strict control during our inspections.
2. Production of the Learning Machine Shell and PCBA Board
There were 80,000 sets for this learning machine project, and they were required to be delivered by the end of June. Our factory had arranged the production plan of the learning machine to meet these two requirements:
(1) Production of Shell
According to the production schedule of our factory, the creation of these 80,000 learning machines would take 1 month to produce. Because the casing of this learning machine was relatively large, we used a large injection molding machine to produce it. In order to complete the production process with the needed high efficiency and ensure our creation was of high quality within the stipulated time, we started all the large injection molding machines in the injection molding department with production focused on this project, and all the incoming materials were ensured to have passed the comprehensive inspection of IQC before they were put into production. While the injection molding machine manufactured the casing, our IPQC was sure to do an inspection every hour, picking out the unqualified products in the production process, and making sure to correct any problems with the needed solutions in the production process in a speedy manner.
After the products were checked to be functioning correctly, we would then engage in the final packaging. Because the quantity was relatively large, in order to avoid the problematic products from being mixed into qualified products during the production process we developed a complete and thorough response plan. That way, once the problematic products were found, they had to be labeled and remarked and put in a box, before they were later collected by the department head and checked by the quality department. When were a large number of orders, we formulated relevant measures to deal with and minimize errors.
(2) Production of PCBA
Because the PCBA board is the carrier for the operation of the learning machine, it needed to be produced according to the circuit board diagram designed by the engineer. The main production of the PCBA board included three processes, namely SMT, chip bonding, and the DIP plug-in. This time the circuit board was exported to Turkmenistan in naked form, and it could face the risk of being struck or become worn during transportation, so we paid extra attention to the production and packaging process. This was especially true during the DIP plug-in, where we would solder the electronic components more firmly and test it several times to prevent the parts from falling off. Finally, we wrapped the packaged PCBA board with foam cotton layer by layer, so as to reduce the friction and collision between the PCBA boards.
Because for this order the products were shipped in a semi-finished state, and the quantity was also very large, the product was a bit cumbersome to safely secure throughout the packaging process, and it was easy to mix different materials together should someone not be paying close attention. To ensure everything went well we focused on making a product packaging operation guide, distributing different materials to different production lines, and ensuring everyone packed and marked the information according to the guide map. By doing things this way, it was convenient to count the number of products at the time of shipment, and it was also easy to sort all kinds of materials.
VI. Inspection and Shipment
- Inspection of the Good
According to the prescribed process, when all product materials and semi-finished parts had been produced, we needed to notify the customer. The customer arranged a third-party SGS inspection agency to come and inspect the product standards. The products could only be shipped if they passed the customer’s inspection standards– should it be found that some products failed this requirement, all products would need to be reworked and re-arranged for inspection again. Because we had made full preparations in the early stage and strictly controlled all materials and products in the production process, the inspection was only needed to occur once for us to pass, helping prevent any delays in delivery.
This batch of goods was shipped in three cabinets, and the number of boxes was particularly large. In order to avoid confusion in the process of loading containers, we arranged for employees to cooperate with each other and transfer the goods to the cabinet. At the same time, we also prepared another plan where should any product be mixed-up we could identify the issue and deal with it at a good rate of time.
VII. Technical Support
While awaiting the arrival of the accessories of the learning machine to arrive in Turkmenistan, the professional technical instructors of our production and engineering departments prepared a number of relevant technical supports to be able to remotely guide our customers when the time came for them to assemble and test the products.
What Was Prepared:
(1) For each job position, we prepared the English version of the corresponding production operation instructions, which was translated into Turkmen by the translator who was on the customer ’s side, for the reference of the workers in their factory.
(2) Our professional technicians, such as the Production Department and Engineering Department, also recorded relevant videos to tell customers which problems they needed to pay attention to during assembly and production, as well as how to solve some machine crashes, etc., and be able to provide remote technical support to customers.
(3) In addition, we also prepared 0.1% material spare parts for customers when the learning machine was shipped in bulk. By having this 0.1% of finished products, they could be used in the case a problem occurred that was not repairable during the subsequent production process, with a certain amount of spare parts therefore being useful.
Eventually, with the mutual assistance of our factory, our customer completed 80,000 sets of learning machines in mid-August, ensuring that every new student could receive a children’s learning machine at the beginning of the school.
This order was different from previous orders. It had strict requirements on the factory’s production capacity and material quality control. From the development of the learning machine software, to the procurement of materials and parts, to the production and shipment of half a year, we relied on our rich manufacturing experience to complete the one-stop process of quality, quantity, delivery, and after-sales. As this was a government project order we not only were sure to strictly control every production process, but we also gave customers sufficient time to complete the assembly and packaging of finished products, and we provided technical support guidance to help successfully complete the order requirements. Through this production we found that even in the face of different types of orders we are prepared and confident in our experience to solve the challenges of any production process.