A new trend at transportation companies: setting up Operational Analysts
Jonathan Bouaziz is the CEO of Kardinal, a tech start-up specializing in the transport and logistics sector. Since 2015, Kardinal offers a real-time tour optimization based on innovative algorithms and strives to become the technological leader in its sector.
Operational Analyst is a little known role in the transport industry while it’s very trendy in some others. However, we can’t say that an operational analyst is nothing but a fashionable role, it has more to do with a natural evolution of logistic jobs.
Jad Naous, VP Engineering & Product at Imply, developed this concept in a meaningful video that I highly recommend you to watch. According to him, automation won’t remove workers. On the contrary, workers will be better assimilated with new missions within the organization. These new tasks will put operational people on high value activities.
At Kardinal, we have always had this vision of the operational analyst but we used other words, like “a tool to help decision-making”. Jad Naous’s video helped me to formalize what I think is an important logistics jobs transformation. In this article, I will share with you in detail my vision for the operational analyst.
In a few years, an operator will no longer carry out most of today’s repetitive, low added-value tasks because machines will handle them. It’s an inevitable and good evolution of operators’ work. This change is already underway in some industries. Think about opinion surveys. Before, a survey manager had to get back little papers, classify and analyze them manually. Now, this mission is done by a tech tool which analyzes and synthesizes data. This tech evolution allowed survey managers to have more time for complex missions. Instead of spending time collecting data, they spend that time acting on it. It’s the same thing in the transport industry with the transformation of the operator’s job.
To understand this operational analyst trend, I will take a quick step back in the transport industry. I will analyze the tech step that led to this logistic work evolution. Indeed, Operational Analyst is part of a global tech transformation in transport companies. Then, I will get to the core of the subject by defining precisely what an operational analyst is. To finish, I will share with you some feedback and tips to implement this new model in your company.
I/ From paper to digitalization, from digitalization to automation
Let’s start with a quick history!
1°) Before digitalization: a tedious manual step
The supply chain was first transformed to constitute a network of warehouses and agencies on the territory. The goal was to collect, select and return the packages more easily. However, the work remained manual. Operators were swamped with paperwork. Order forms were exchanged thanks to the telephone box then over fax. Operator tasks had a low added value. Who is nostalgic for this period? Passing information it was an oh so slow process…
2°) A digitalization in two steps
Then came the era of digitalization. Thanks to the Internet, interactions between operators became smoother and simpler. Time treatment had been reduced considerably: documents were sent by mail and signed digitally; customer orders were made online. It was the first step of digitalization.
Afterwards, operators started to establish processes to define the company’s micro and macro needs. Supply chain consulting firms popped up all over and internal competencies were developed to redefine the delivery process or the services being offered. It was the Business Intelligence period: data used to answer questions but it took time! The waiting time was several weeks.
During this step, data became a goldmine! The tools of Business Intelligence, Transportation Management System, Enterprise Resource Planning have been installed to leverage this new resource. Intelligence has been put in processes. This tech evolution and the beginnings of automation have had a large impact on work operators: it gained in added value!
3°) The dream of a complete automation
This step of digitalization opened the way to a complete automation, which is still in the air! This dream came to life through the automation of some tasks like the printing labels for the packages, the automatic package sorting or automatically offering a delivery time slot to customers.
Yet, this process of automation is more or less successful. Take as an example the automatic package sorting. The machines are very expensive and they don’t necessarily recognize all addresses, handwritten or not. As a result, these systems aren’t always that profitable. But it’s not as simple as that! If standards are set up, the automatic package sorting becomes money-making and saves time!
Automation can be an achievement. Think about the automatic appointment-making which works and leads to a productivity gain. But it can also be a mixed-success or a failure, like the delivery of packages by autonomous vehicles which isn’t conceivable yet.
Concerning the complete and extreme automation, I think it won’t work. People think that by removing workers from the supply chain, there wouldn’t be any more mistakes. But it’s nonsense to think that. For it to work, data should be perfect, “clean”. However, from my experience, that isn’t the case!
Digital Supply Networks: Transform Your Supply Chain and Gain Competitive Advantage With Disruptive Technology and Reimagined Processes (Amit Sinha , Ednilson Bernardes, Rafael Calderon, Thorsten Wuest)
And there is another factor which must be taken into account: a part of the field knowledge is in the operator’s head. Here’s a meaningful example: an artificial intelligence can’t know that a delivery driver argued with a customer and that these two people mustn’t meet again. But, currently, this day-to-day field knowledge isn’t incorporated into an organized process. And I don’t see any IT system integrating this data point anytime soon.
The best bet to make is to leverage human knowledge. With the automation development, time has been released for operators, so now is the perfect time to take advantage of operational high skills! But how do we leverage this knowledge? How do we give it value?
II/ The operational analyst: not a new job but a transformation of logistic works
The Shift in Day-to-Day Activities thanks to automation: people are becoming Operational Analysts (Image Credits: Jad Naous)
Take two examples!
A branch manager knows how many trucks and assistance resources he needs for Black Friday. For this, he analyzed last year’s Black Friday and his Excel table with a lot of approximate information, then he fixed a number. However, this way of thinking isn’t reliable because his scope is always moving. Information can be correct now and flawed in two minutes… Thanks to data, he will have a better and more rational management of his needs.
It’s the same thing with a trucking manager who defines the delivery time slots: his planning is based on a typical day and his experience. His approach is biased: a route can be perturbed by a lot of factors, like a traffic jam. Thanks to route optimization, an operational analyst can react and find a solution quickly. A tech solution provides more fluidity.
Let’s clarify our definition of operational analyst: it’s an operational manager who uses data to make better decisions and his decisions have a direct consequence on C-Levels and drivers. It’s a central and important actor in a transport company.
But let’s go into details!
2°) On what and how they work?
An operational analyst is focused on decision-making. For this, he leverages on:
- His field knowledge,
- His knowledge of different goals, those of the company, the agency and the drivers,
- His central position in the organization,
- Data, which is his main material!
These elements allow him to make contextualized decisions (instead of generic or biased decisions).
Imagine an operator who wants to make a driver leave 10 minutes later. He must be able to determine all impacts of his choice, on the fleet of trucks, the customers or the productivity. Optimizing a route is a complex mathematical problem:
- When a route includes three addresses to deliver, there are six possible routes,
- When a route includes ten addresses to deliver, there are 3 628 800 possible routes!
Without artificial intelligence, it’s impossible to have this global and scientific view!
Another situation: a client changes his delivery time slot at the last minute. If the organization manages not much delivery time slot – 5% for example -, it can deal with this change manually. But if the percentage is 50 or 60, the modification isn’t manageable without a tech solution.
Better, with an AI, an operational analyst will be in position to test new scenarios because he will know the consequences immediately. Before, he had to undertake an important research or use classical formulas (often biased) for lack of time. AI produces more creativity!
But data is nothing without human knowledge too. Let’s be precise: it’s the operational analyst who makes decisions and not artificial intelligence. The solution is an assistant, a kind of alter ego, which calculates all financial and organizational impacts of the choices taken by a human. And, at the end, it’s the operational manager who has the last word because he understands better than the AI the consequences of these decisions on all organization levels (From drivers to C-Levels). Organizations are very complex, the interactions between operators are increased and the information is always changing. Then, an operational analyst uses the combination – knowledge and data – to decide. In a way, he has two allies: data and his high skills!
Let’s go back to our example: Making a driver leave 10 minutes later means that a customer will be delivered a few minutes late … But if this choice increases global productivity, is the operational analyst ready to take the risk of not satisfying a customer? Only a human can make a balance between several goals. In our example, he will answer a business objective that an AI doesn’t know. In other words, an AI generates precious algorithms, but it isn’t capable of making the right decision in some situations. A machine isn’t aware of all the consequences of its decision.
III/ How to support operational analysts in practice
1°) Deploy new decision-making solutions in their hands
The logistic jobs transformation and the setting up of AI are important evolutions which aren’t installed immediately in a company.
To manage this transition in your company, here are 4 tips based on my experience at Kardinal helping transportation leading brands:
- Operators need to understand the stakes of this transformation and data significance in day-to-day work. In front of experts, you must be able to prove the Return on Investment of your tech solution. Your arguments must be quantifiable.
- The operators will challenge your solution. Then, pay attention to their feedback. Their inputs are very important: you will include their constraints in your solution.
- To set up a methodology which allows incorporating the tech tools in daily work operators. Your operators must have faith in your solution and consider it like their assistant. A structured methodology is the key. If it can help you, in an article, I describe the process devised by Kardinal.
- Be aware of different perceptions. An operator may feel that the solution isn’t working while the results prove the opposite! When you offer a new solution, don’t forget to come back to metrics to show that the AI is performing.
2°) Ensure solid data to answer real-time needs
Your data must be solid for your route optimization to work. Your operational analysts want to obtain an answer quickly to react to situations like:
- New and critical situations. Covid-19 is a good example: the B2C volumes increased considerably and, paradoxically, a lot of transport companies lost money. Why? They hadn’t enough resources to answer demand and manage the shock.
- Volatility of volumes, for example the famous customer who makes a mistake about the weight of his package to pay less for his delivery.
- Variety of products, like a tire which must go inside a truck.
- The organization always moving with a lot of interactions and administrative decisions. For example, an area closed to trucks which will affect the organization and the productivity.
Your operational analysts have a need for immediacy to determine fast financial impacts of all these complex situations. And data answers these new operational stakes. If your marketing team develops a new offer – for example a two-hour time slot -, how will you manage this new offer operationally? What will be the impacts? Should you have a new fleet dedicated to this offer? Data will analyze all scenarios and bring measurable replies. When your margins are weak, you need numbers!
If your company doesn’t support the logistic jobs evolution internally nor set up a tech solution, your quality service, productivity, organization and profitability will deteriorate over time.
To drive growth at your transport company, I recommend you adopt this new approach in logistic jobs, as it is more rational and scientific. Focus on the automation of repetitive tasks first. Once you’ve completed this step, you will be able to implement a tech solution in your company and your operational analysts will focus on complex missions.
At Kardinal, it’s the vision that we have of the transport industry evolution. In my next article, I will say a little more about our approach!
Jonathan Bouaziz is the president and co-founder of Kardinal, a company that publishes a solution dedicated to the optimization of the last mile. A logistics expert with a passion for technology, he dedicates his expertise to the implementation of innovative solutions to build a supply chain in line with the new challenges of his market.