While we cannot quantify everything that we are capable of doing, we hope to at least demonstrate our skills through some of the past projects that we have done.

As you explore more of our services and offerings, you can begin to see how we combine our skills and capabilities in some of our projects in order to help our customers achieve the results that they required.

With that, here are some of the use cases.

The client, Diyou Fibre (M) Sdn. Bhd. (DF), is a fibre manufacturer who was trying to track their work-in-progress (WIP) stock at multiple stages of their production process.

While tracking WIP goods is slightly more straight-forward if the goods were in constant production flow, this was not the case for DF. In fact, their production was based on the fluctuating demand of their clients. To make things more complicated, DF would also sell some of these WIP stocks as soon as there's a willing buyer.

(Image above is for illustration purpose only. Original flow has been modified for confidential purposes.)

SmartB was given the task to help track (and trace) the WIP stocks, starting from the raw material stage, all the way till the goods leave the warehouse. Stocks that are kept in the warehouse will be traceable too.

The WIP stocks are usually kept in a bag (weighing between 200-300kg) and simple records of its production needs to be tracked as well. There is 3 warehouses where the goods are kept.

Lastly, most of the staff involve in the production are foreign workers who cannot speak or read English.

The first part of the solution was to figure out how to make we captured data conveniently and accurately. Using keyboards and mouse would not work as that is more suitable for office or administrative type of operations.

Hence, the answer was definitely to use touch screen. While a tablet would work, a touch screen terminal would be even better. In fact, most of the terminals that we sourced are designed to be heavy duty but at a reasonable cost.

The customer wants to track their goods. Since these are manufactured in mini pallets, the best way to keep track them was to have weight record with each bag.

The bag in the image above is a similar type of bag used on DF.

As part of the move towards paperless, we manage to integrate the weight reader with our terminal. Hence, allowing the client to record the weight of the bags as they are produced. This would also instantly record the time of manufacturing each bag and, indirectly, the production rate.

Another thing to note is that our system is a cloud base solution. Hence, all data captured are instantly accessible online.

As the production workers are predominantly foreigners, we had to design a dummy proof system - big fonts, big buttons and limited words.

Again, with some basic training, the system was merely one that requires a few click of buttons.

A challenging project where SmartB had to utilize many unconventional ideas before narrowing down the options to create a suitable and practical solution.

As shown above, this was one successful and creative project delivered for a very unconventional tracking case.

(Legle is a brand fully owned by Inhesion)

The client, Inhesion Industrial (M) Sdn. Bhd. (IM), is a tableware manufacturer who was trying to implement an end-to-end ERP system which predominantly include the Sales, Product Development, Production, Inventory and Accounting modules.

This use case is to describe a custom-made module for their Product Development management. As shown in the image above, Inhesion specialises in custom made and custom design tableware.

For this project, SmartB worked very closely with IM for several months to first identify the requirements before building the modules.

(The image above has been kept slightly blurred for confidentiality purposes)

IM has sales representatives in various parts of the world. The diagram above shows the process of each new request submitted to the production team and how the approval processes were made right until the birth of a new product for sales.

All there were important because these new products, when developed, need to ensure that all production cost, quality and profitability are assured and within their expected range.

SmartB's challenge was (1) to understand the end-to-end process before (2) ensuring the feasibility of the design that was being built. Lastly, (3) to ensure the good user interface (UI) and user experience (UX) were achieved.

Essentially, and metaphorically, SmartB was like the architect and the builder at the same time.

As this was considered a mid to large size project, we broke it down into sizable chunks. Essentially to deliver it by multiple phases.

The goal was to immediately digitize and streamline the process. Hence, no more physical forms or even emails allowed.

With that in mind, the first thing we built was an digital request form where any requester would need to fill in the request form online.

As soon as they submit the form, the request is immediately logged into the database.

Once the request has been submitted, the person-in-charge (PIC) is next to take action.

As you can see, the request needs to be managed and a simple dashboard was built for that. Multiple tabs were also added as this were different parts of the request.

With all the activities involved in the Product Development - which we identified during the initial need analysis, we had to build an request engine.

For simplicity, this means we had to build a module to initiate and take action the various process logic.

For example, when a request is submitted, the requester would have given a cost allocation breakdown for the various departments. Using this breakdown, relevant head of departments (HODs) would receive a notification that would prompt them to review the request.

Of course, this is just one of the many features.

After we built the monster, we now need to tame it.

As part of the design work, we decided to only show fields and tabs that are relevant to the specific request. That means, only 50-70% of the entire system is visible to the user at any point of time. Non-relevant fields and tabs are dynamically hidden from the user.

A 100% custom-made system was built from scratch.

Thanks to the strong emphasis made on the design and planning stage, the new system was delivered - meeting nearly all the requirements made by the IM.

Furthermore, building this system on top of our standard modules also mean that IM can further benefit from the existing system. A synergistic effect where 1 + 1 = 3.

The client, V'Asia Cosmetic & Healthcare (VA), is a health and beauty brand owner who was trying to implement an end-to-end tracking system.

This is important because, as the VA becomes a more reputable brand in the marketplace, they are faced with integrity and social challenges. Some of these includes:

• Mis-match of data between VA and their distributors
• Product traceability, and
• Product authenticity claims

Also, VA is no longer a small company, without a proper tracking system, some of the operational challenges can be quite costly.

The client wanted a system to trace all stock movement starting from where the goods were manufactured.

That means, the data capture of the end-to-end flow needs to start at all the manufacturing sites. The client presented us with 5 manufacturing sites which were predominantly based outside the city.

However, as some of you can imagine, not all locations/site has good internet connection. This issue alone made the project extremely tricky.

The process flow for this task is probably a simple one where the factories produces the goods, they box the goods up (in bulk) and dispatch it to VA's warehouse. As soon as there's a customer order, VA would pack the necessary order for further dispatch.

Hence, the real challenge lies not in the complexity of the process but in the complexity of the implementation and set up.

Hopefully, by understanding these challenges, it illustrates the level of difficulties faced.

In line with the above, we worked closely with the client to identify strategic locations within each factory in order to ensure that

• The internet signals were stable
• It is convenient for the production team
• Data collected were relevant

Another challenge was due to the fact that, for hygiene purposes (for food manufacturers), some of the factory were insulated. Hence, internet signals were weaker at certain points.

As the stickers were distributed randomly to various factories, there was no way we could to identify the serial number used in any production plant in advance.

Hence, we had to figure out a way to register the product to the relevant serial number (through the QR Code). The somewhat small packing area (for some plants) also meant that we had the following restrictions:

• This must be done as soon as the sticker was pasted on the product
• It must be done quickly as the product moves along the conveyor belt
• It must be done immediately before the products were packed into its box

Meanwhile, the system would also group that specific range of serial numbers to specific boxes.

Lastly, with the information from the product profile, we had to instruct the system to print the box labels according box capacity. E.g. if the box could only accommodate 30 bottles, the printer would print a new label as soon as the 30^th QR sticker was scanned.

That way, we could identify the range of serial numbers moving in that specific box each time we scanned each individual box.

Looking at the label above, you can see that the range of serial numbers are displayed on the box label.

Because this is a semi-automation process, some of the warehouse staff still needs to visibly see the range of serial numbers use for that specific box.

Imagine this, you have 300 boxes arriving in the warehouse (from the factories), each box holds 50 units of health packets and you need to register the serial number of each item entering the warehouse in less than 30 minutes.

In other words, we need to know which are the exact 15,000 serialize products that has entered the warehouse in that duration. Yet, the production was only a semi-automated process, the WIP goods are moving as fast as the conveyor could move but tasks were predominantly executed by manual labour.

That was the task assigned to SmartB.

The first thing was to integrate a wireless Bluetooth scanner to the system. Looking at how a operator works, we estimated that he/she would be able to scan around 5 boxes in 3 seconds. Using that information, we identified the right amount of information captured during the scan and automated the rest of it via the backend system.

Yes, it was successful task.

Majority of the goods are sold in boxes. However, there are some goods that are sold as loose items. Meaning, while the box might hold 60 bottles of health supplements, some customers might only want to get 10 samples.

Meanwhile, you can’t expect the warehouse to scan the remaining 50 bottles if they sell them all at one go, right? Hence, the task was to be able to still scan the box label and know the 50 serial numbers that remained in the box.

As you can imagine, this would have been the ultimate goals from the start. However, when combined with all the challenges above, the system had to be modified in order to achieve the same results.

SmartB successfully overcame the obstacles that came with the factory environment as well as some challenges from the above average processes that was in place.

This is important as many implementations in that has failed (in the market) not due to the capabilities of the systems but due to the inability to overcame challenges on the ground.

The client, Inhesion Industrial (M) Sdn. Bhd. (IM), is a tableware manufacturer who was trying to implement an end-to-end ERP system which predominantly include the Sales, Product Development, Production, Inventory and Accounting modules.

Use Case 2 (see above) was also a project for Inhesion. However, this is for Inhesion Asia (IA) which is predominantly the commercial/ sales team.

IA is a Hong Kong based company. However, they have they have an extensive worldwide sales channel – which covers Hong Kong, China, France, Macao and America. With such a large sales network, they needed a system which could provide the following information conveniently.

• Visibility of stock availability (to determine the lead delivery time)
• Price (for various regions HKD, MYR, USD and etc)
• Controls and authorization (from various Head of Department) – for changing prices and for tiered pricing model
• Production information – for specific sales channels
• Product information – for all visitors

All the above information needs to be provided for each individual product code (or SKU).

(Diagram for illustration only)

The diagram above is a simplified matrix of how the information is accessible based on:

• The users’ level of authority
• The region where the user is registered
• The type of product information

In order to resolve this challenge, the first thing we had to do was to break down the requirements.

Similar to the project which we implemented with VAsia, each QR Code is displaying a URL link instead of a product ID (or code). When scanned, you will be directed to a web catalogue that would look similar to this (below).

Here, all viewers can see the generic information of that specific product. At this stage, all you need is a mobile web browser and any one can view this page.

In order to manage and restrict the information accessible by the individual, all users will have a pre-defined user profile. Essentially, the profile is based on the level of accessibility they should have and the region (or country) in which they are in.

(The image above has been kept slightly blurred for confidentiality purposes)

We successfully created a mobile app that reads new product QR codes on-the-go and it is integrated with:

• Sales information
• Production information
• Inventory information

With that, it means we can display all the information required by the sales team when in front of the client irrespective of where they are in the world.

The client, Taiko Marketing Sdn Bhd is the only supplier in Malaysia to produce Chlor-Alkali AND Sulphur Derivative chemicals.

And they were trying to (1) optimise their production while juggling to balance between short-term fluctuating production demand, (2) managing production priorities and (3) managing manpower overtime cost.

In short, it was as if they need an Artificial Intelligent (Ai) to tell them how to prioritise their production orders. Unfortunately, most ERP systems would not have a module that is suitable for Taiko, while those that may have it, usually ends up costing too much.

Apart from having a tight budget, Taiko needed a SmartB to help them formulate the algorithm to significantly reduce non-production activities (e.g. switching and/or setting up new machines) while regularly updating the production priority based on short-term urgent requests.

If this was not hard enough, it took Taiko a significant amount of time to conceptualise what they needed as this module was very new to them. Basically, they had a vision of what they wanted, but didn’t know how to explain it in simple terms.

Note: In order to maintain confidentiality, the information in this case were altered. However, rest assured that the complexity of the requirements remains at similar levels.

In order to resolve this challenge, the project had to be implemented with a sequence of activities.

The first part of the project was to help the client clarify the parameters of the system that they wanted.

This includes listing down as many “what if” scenarios as possible.

As more scenarios were identified, new potential challenges/ outcomes started to unveil itself. More importantly, we had to rectify some of the issues from those outcomes, and decisions had to be made in terms of controls and management of those outcomes.

The real benefit of modelling the system was that it gave both the client and SmartB an opportunity to see the “blind spots” that we would not have envisage unless we saw the model in action.

From that, we generated supporting databases to ensure that the main algorithms would function as planned.

As the project was complex, the need to remain focus was crucial. So, instead of building a "robot", the idea was to focus what was important.

That means, we focused on optimising the production sequence only. And that would have meant that the bulk of the requirements would have been met.

Although no AI was built, we could still achieve a fairly high level of success.

In order to maintain relatively low cost, a MS Excel model was built in order to mimic the desired system. To a certain extent, we were prototyping the potential system that was required.

In order to mimic an ERP system, MS Excel Macro and Visual Basic for Applications (VBA) were used.

Note: Explanations were given at every stage in order to ensure that client was aware of the difference between what was shown in the MS Excel and how the system was function (if one was built).

The end result was that the client started to use the MS Excel as part of their daily operations. The model was a success as it provided a clear structure of what was required in the event a system was built.

The following video shows a section of the entire model.

Note: At the point of writing this case, Malaysia is going through a Movement Control Order (MCO) due to the Covid-19 virus. Hence, the system was not built yet but the model is already in use as a tentative solution. 

Note: A Some parts of the video have been blurred deliberately in order to maintain confidentiality of information. 

The client, LTI Resources Sdn Bhd (LTI), is one of the leading self-adhesive tape manufacturers in Malaysia with in-house R & D, design and converting facilities.

LTI approached us to help them optimized their cutting sequence as the current process incurs a high amount of wastage.

In order to explain their challenge, here is a simple customer order that will, hopefully, illustrate the problem - Note that these have been modified to maintain confidentiality.

LTI receives a mix order from a regular client for the following:

• 12mm x 50m = 100 rolls
• 18mm x 50m = 120 rolls
• 24mm x 50m = 200 rolls
• 36mm x 50m = 110 rolls

LTI currently has the following available stock:

• 200mm x 50m = 150 rolls
• 130mm x 50m = 30 rolls
• 70mm x 50m = 15 rolls
• 65mm x 50m = 3 rolls
• 1020mm x 50m = 3 logs (Full Size)
• 750mm x 50m = 1 log
• 1020mm x 1000m = 10 jumbo

Which existing stock should LTI use to ensure that there is minimal leftover/waste? And how should LTI combine the size when they cut the rolls?

In order to resolve this challenge, the issue needs to be solve using mathematical logics. Here's a visual illustration of how we can solve the problem:

(Image from Google)

As shown above, in order to optimize the cutting process, an intelligent system was developed in order to find the combination of sequence that produced the least wastage.

The end result was that we built a system that could automatically calculate the best cutting combination and this could be done within seconds.