We sat down with Komatsu's Mining Technology Solutions team, led by Shingo Hori, Senior Vice President, Mining Technology Solutions and Anthony Cook, Vice President, Autonomous Mining Systems, to discuss how Komatsu helps partners transition from manual to autonomous operations.

Q: How early should companies be thinking about change management in regard to AHS?

SH: The journey starts with understanding the need for change, and whether the change to AHS is the right sort of change. Ideally this process should start before any sales interaction happens. It starts with us engaging with the customer, where they understand who we are and we understand who they are.

AC: When we sit down with the customer, we're interviewing them as much as they're interviewing us. We try to get the customers to ask themselves what problems they are trying to solve with AHS. For example: What is the main reason you are doing this? Is it safety? Is it productivity? How are you going to support this? Do you have the technical knowledge, the operational knowledge?

Q: What needs to go right to successfully transition to autonomous?

SH: The business factor is important, but the standout factor is the relationship. Implementing an autonomous system is about going on a journey with the customer. It's about continuously improving together with the customer to continue to provide them with the safety and productivity benefits. This requires a fundamental change from the old supplier-customer type of relationship to a partnership.

The most successful deployments have been where we've had access to the site well in advance - understanding how the customer operates, where the operational pinch points are, and whether FrontRunner is even the right fit for the challenges in their organization. Having an open and honest conversation at various levels of both organizations is the key to ensuring a shared culture of change.

AC: You can't just order AHS on Amazon and it turns up and you install it. It's a multi-year process just to get it off the ground. There will be challenges, and you need to have that trust and fundamental relationship with each other so when these things happen, you know we're working on it.

Q: Are there ever cases where AHS isn't a good fit?

AC: AHS is certainly not a panacea for all that ails a mine. We have seen mining houses that approach AHS as just a procurement exercise. Instead of looking at it as a production tool and an enterprise-level process tool, they look at it as if they're buying hardware, another dozer or grader or whatever. Those are the ones that are never able to get their projects across the line, because they're not approaching it with a fundamental understanding of how AHS can potentially change them as a mine operator.

SH: There are a number of physical elements that come into play - for instance, the system has to be able to see a certain number of satellites, so if it's an operation that has significant highwalls or visibility issues - it presents a problem. FrontRunner might be absolute overkill for a small mine having problems with operator driving behavior. More often than not, semi-automation or operator assist technology is often enough to get them to where they need. But from my experience I can't recall an instance where we've declined to offer because of the physical nature of how the mine is constructed. It has mostly been related to practices, cultural alignment - the soft side of things.

The important thing here is our objective is not to sell automated systems. We want to sell the solution that is best for our customer. If FrontRunner is the right solution for the challenges the customer is facing and we mutually agree that's the case, then it's our job to make sure it works.

Q: Does any extra work go into deploying in places with extreme climates?

AC: It can - we had one mine site with extremely adverse conditions at times.  Our development division led a project where we deployed two trucks over the course of the four seasons. We targeted that period of the spring breakup where you get massive snowmelt and all the roads turn to slush. We measured very low-level details like torque and performance and from that we understood the behaviors of the truck. The next season we made modifications to hardware; there were things like mud-flaps going on to prevent debris from splashing up and covering radars. Once we achieved a base level that we were happy with and thought the customer would be happy with, we did things like intentionally running the truck on the same path and getting it bogged, to see what would happen and how to get it out. We implemented this in a trial phase and now we continue to monitor. It becomes part of the ongoing roadmap of improving truck performance.

Q: How much time does it take to get from a decision to having the autonomous operation up and running at the mine site?

AC: The sales cycle process can run nine to 12 months. This is largely because once we've provided indicative pricing, our customer has to build the business case and present it to their board. After that, it typically takes nine months to get a truck or kit to the mine site. A lot of work goes on with the customer during this time. This includes figuring out the intricacies of the support contract. When we start out, Komatsu takes on more of a role to get the customer up to speed than it would in a normal manual operation. As we mature, we pass on roles to the customer so they become more self-sufficient.

Q: How long does Komatsu maintain involvement with the customer?

AC: We never completely detach from the customer because if we do, we start to lose the ability to evaluate and support the product properly. We know there will be times when things go wrong. It's not like one truck going down; if the whole system goes down then there is no production, there is no money coming out the end of the pipeline. Things have a tendency to escalate very quickly, so it's really important we have those lines of communication in place.

Even with our mature customers, we still have very regular engagements on productivity improvements. For example, we might run a Pareto analysis of what are the delays that are causing the most impacts to productivity. We'll scope those out by asking: what are our top five delays? It could be false alerts or people hitting the E-stop button by accident; whatever it is, we'll focus on it as a training issue.

We are establishing an order of priority for each request for development. Safety gets the highest priority, obviously. Then we'll look at productivity impacts. Any time we can improve speed or get the average cycle time reduced, we're making improvements to the customer's bottom line. It's a continual evolution.