IoT in Logistics — My Experience

The growth of logistics in India has shown tremendous improvement in the last decade, starting from scratch and reaching a level where the Indian logistics industry is competing with the top dogs of the most affluent countries. The industry itself was valued at $160 billion in 2019 and was on course to reach $250 billion this year, before the pandemic set it back. It will soon be the most sought after industry in India.

Ok.. so what’s my point?

While it is doing substantially well, there are areas that definitely call for a review and improvement, mainly in terms of infrastructure and advancement in technology.

Today i’m here to write about my experience on the latter; Specifically — the introduction of IoT in Logistics.

At my previous company, Decathlon, my role required me to oversee the entire transportation operations within India. I was specifically in charge of outbound movement, from our warehouse to all retail stores.

Working in logistics, especially in India, is a bit challenging with number of restrictions and obstacles in place. It’s typically an industry where you find order in chaos.

After learning the ins and outs of the industry, I wanted to bring about some change to our final mile delivery through the introduction of a ‘click and track’ software, where there would be 100% visibility of every shipment to all stakeholders involved.

My role in this project initially was split into two — one was to ascertain the requirements of all stakeholders in Decathlon & the second was determine the transport provider’s requirements. After these were defined, I arrived at two goals — the software needed to show all the information under one dashboard which was easy to scan and analyse by each stakeholder while the hardware needed to be easy to manage with minimal intervention from the transporters.

In terms of software, we had little to do apart from defining the information architecture as per our requirements and since we were dealing with a third party company, we were provided with a dedicated team to provide a customised user experience. But in terms of hardware, we had a variety of devices to choose from and it was crucial to make the right call. I decided the best course of action was to install dedicated devices (permanently attached) for our local milk runs and short haul deliveries since we used the same vehicles over and over. But for long haul trips, we decided to deploy the portable trackers which required no intervention from the transporter side (the device was just required to be inside the vehicle at all times, with enough battery and a signal).

Redesigning a system in a relatively rigid industry was a little tricky as I needed to make alterations to how people go about their daily activities, which meant complete automation in the documentation process, lesser communication between parties and more autonomy. Moreover, everyone needed to understand the positive impact this would bring, to gain the maximum potential of the software.

Example of information flow using IoT

Who are the people involved in the process today?

1. Warehouse — The pickup point. They are in charge of the products being picked, packed & shipped and ensuring all the paperwork is in order to avoid any roadblocks along the way. In my experience, the majority of problems that were present was in the organisation of paperwork, especially for long haul trips. These details come under the microscope when vehicles cross state borders and if a single detail is found out of place, a heavy penalty would be levied.

2. Retail Stores — The delivery point. Replenishment of goods is their only concern — if they don’t have the products, they have nothing to sell. As a result, they need to ensure it arrives on time, with the right products and qty. The maximum dispute here is when the shipment is on route. There is constant back and forth communication with the warehouse and the transport provider, on the location of the vehicle and ETA.

3. Backend team — In charge of analytics, troubleshooting errors, compilation of data and transporter performance to push data driven solutions.

Ok. What’s so hard about implementing IoT then?

Bringing about change in an industry, especially in the form of technology, require people to alter their mindset and having the willingness to change. In this scenario, I found that in spite of all the pros, the skepticism made everyone revert to their old ways after our first test run.

What I found challenging -

1. Trusting the data — After implementing the software, it was inevitable that most stakeholders had doubts on the accuracy of the information being relayed. Hence, in spite of having the technology, the constant communication between teams continued as usual.

Although, as it became more routine, people slowly started trusting the process (after 6 months of implementation)

2. Reverse logistics — Since the software and devices which we used were owned by us, we needed to ensure it was being circulated without any hassle to ensure availability. This meant the transport providers needed to ensure the device is dropped off at the final delivery point along with the cargo and the same needed to be sent back to the warehouse. Since it was not financially feasible to have a new device for each trip, the reverse logistics of each device was crucial to ensure that each and every vehicle was equipped.

Eventually, the responsibility was handed over to our transport providers purely down to their vast network throughout the country. It was more efficient in terms of cost and process for them to manage the reverse logistics of devices and ensure it was in place before each trip.

3. Communication — Every now and then, issues were bound to arise in terms of a device not working especially when its on route. In this scenario, the driver would need to communicate the issue accurately for it to be resolved and would need to rectify the problem once a solution is provided. This back and forth communication was inefficient and time consuming.

In all honesty, this is very much a problem even today. Efforts have been taken to limit the number of issues in the hardware. But there is always an odd case which pops up..

4. Driver Training — One of biggest challenges, which we overlooked, was the training of the drivers. Yes, the devices deployed needed minimal intervention from the driver yet it required supervision to ensure that it was transmitting data. This includes ensuring the device was charged at all times and receiving signal as well. But what we failed to consider was that most of the trucks which we hired were ad-hoc vehicles, meaning that we had no control over the trucks which were hired from the market. Essentially, we would have to train the driver each and every time a shipment is planned. Not efficient!

For the first six or so months, we had tested a few different devices for our long haul trips and eventually narrowed down to the most basic device which had only an on/off switch to it. The driver’s only responsibility was to check if the device had enough battery through the duration of the trip.

What worked then?

1. Analytics — The moment we launched the first vehicle, the advantages were evident. The sheer amount of data we were able to analyse was invaluable, in terms of truck location, dynamic ETA’s, no of stops along the way, traffic analysis, automated documentation (Product checklist, Lorry receipts and even inter state permits) etc. Moreover the customised dashboard provided each stakeholder all the information pertaining only to their requirement.

2. Time — The most precious resource in the logistics sector. We saved time. On an average, the time taken to dispatch an outstation vehicle reduced by 25% and 15% for local deliveries. This was mainly due to the automation in generating documents. This directly correlated to lesser number of discrepancies at state border checks due to lesser manual intervention.

3. Route optimisation — Since I was implementing this system for logistics in retail, the final customers were the retail stores around the country. This meant each movement to a store was based on pre arranged schedule. Once the user inputs the details for each route, the system automatically generates the approximate lead time (taking rest stops into consideration) and fastest route available. This was again dynamic and could change based on traffic or any road blocks along the way, allowing the stores to plan their activities more efficiently.

4. Alert Management — One of the most important features — all stakeholders would receive alerts on issues that arise during the shipment. For example, if the vehicle is idle for an unusual amount of time or veers off the usual route programmed in the system, low device battery, dynamic ETA changing based on delays an alert is sent to every team in the loop. This eventually minimised the amount of communication between each person within the organisation.

This transformation within the company took almost a year and a half since implementation, to streamline. It required a large investment and a fundamental change in the way logistics functioned. Of course, there were a lot of ups and downs during the process, but no-one could discount the value IoT had brought in logistics especially when there was direct impact on efficiency.

Tech is clearly an enabler and with everyone providing similar features, experience design is what will set players apart.

More on this in my next article..