In Part I, we started using Goldratt’s six questions to judge how technology might help us better control contracted resources at a mine site. Our ambitious agenda required the solution to include group functions and vendors along with the teams directly involved.
[ Listen to audio version, read by David Hodes]
This is Part 2 of our series on Six Technology Questions: Part 1 | Part 2
We not only wanted to control costs but also wanted the system to integrate with the work management software. This integration would enable a real-time understanding of the current and future workload measured against the available resource capacity. We hypothesised that knowing load and capacity would allow us to make solid business decisions, not predicated solely on controlling costs, but also understanding how resource availability might impact schedule. After all, a day of lost production usually dwarfed the cost of holding a few extra resources in reserve.
As a recap, here are the six questions again.
1. What is the power of the new technology? What are the boundaries where it will and will not work?
2. What current limitation or barrier does the new technology eliminate or vastly reduce?
3. What policies, norms or behaviour patterns do we use today to bypass the limitation?
4. What policies, norms and behaviour patterns should we use once the new technology is in place?
5. Given the above, what changes and additions to the new technology should we introduce?
6. How do we integrate the new technology with the overall strategy?
In Part I, we looked at the first question (actually a pair). Now let’s see how the other questions combine to give us the ability to tease out our real requirements and judge any proposed solution against these criteria.
The most fundamental limitation the technology eliminated was the answer to the question: ‘Who do we have available to do the work?’ Once implemented, we could get past the embarrassing situation of having to ask the vendors which of their staff had turned up to work on which project.
But that was just the beginning. We could tell who was supposed to turn up, whether or not they were ‘good to go’ in terms of quals and OH&S, and what we were going to pay them based on their contracted schedule of rates. We could report in real time what the hours and money were by project, by resource, resource type and by vendor.
As the single source of truth, we were able to provide the vendors with the data they could use to lodge a payment claim without any further reference or checking with our client. Any variance they had, with authorising overtime, for example, could be resolved by exception.
“A day of lost production usually dwarfed the cost
of holding a few extra resources in reserve”
We were able to report on absenteeism both on the day and as a trend by vendor, person, and resource type and were more than a little surprised to find that in some instances, attendance could be as low as 60% of those people forecast to be onsite on any given day. Having the ability to measure attendance against the demand for resources as dictated by the schedule, we were able to collaborate with the vendors to ensure we mobilised critically constrained resources well enough in advance, and thus proactively mitigate costly schedule delays.
With the rollout of the technology, it became possible for the first time to know at any given moment, who was onsite, with what skills, doing what work. Also, we had in one database the aggregate demand for all work across all projects for any slice of time, in real time. Through the use of Critical Chain, we were able to use buffer management to intelligently assign resources to those places where they would have the maximum impact across the whole project portfolio.
The foundation of the old ways of working was the sign-on sheets used at the prestarts. The assumption was that the people signing on were ‘good to go’ with all their quals and OH&S requirements. But, given the flood of people coming and going, the only time the team validated the quals was either when a person started for the first time or when they were first assigned to a given project.
Thus, the norm for vendor administrators was to receive downloaded CSV files from the gate-access system and reconcile them with the manual spreadsheet tally of every prestart sign-on sheet. Whatever the vendors came up with as their charge was usually accepted by our client’s controllers, provided the hours of attendance reconciled. It was beyond practical endeavour to check every person by craft, rate type and shift penalty.
A further accepted norm was based on the fact that the designers of the ERP system configured it with different master data for resource types to that of both the vendor craft and the Learning Management System (LMS) classifications. The controller team resigned themselves to the sad fact that the only way of analysing and reconciling was to do it manually through spreadsheets and lookup tables.
A behavioural norm built into the calendars of the two sets of administrators was a weekly meeting to go through all of the disputed records. Not only was this work tedious for all parties, but it often meant interrupting coordinators and supervisors in the middle of their work to answer questions about, for example, overtime and work completed offsite on behalf of the project.
Having little idea of who was coming, or actually onsite, on any given day meant that resource allocations both within and across projects became a function of whose wheel squeaked the loudest—not usually the best way to achieve the optimum business result.
Once we had built the technology, our clients were able to create open positions for specified project roles, based on skill types and rosters. We enabled the vendor administrators to populate the assignments for the published open positions, and we could validate, through the LMS that the people they were offering as candidates had all the necessary quals and were current with their OH&S mandates.
Policies were put in place to recognise what constituted overtime. It seemed so simple to introduce a business rule that could be managed by computer, that said, for example, that the billable workday started at 6am, regardless of how much earlier the worker clocked on.
We advocated for the idea that the best results are accomplished when all parties see the win-win. Along with the health and safety inductions, we introduced some elementary training about the end-to-end process, from pit to port. We highlighted how the projects they were working on acted as a bottleneck to production and how, if they performed their work to time, budget and quality, they would make their contribution to the overall goal of safe, reliable production.
Once we had deployed the baseline requirements of the new technology, we turned our attention to how we could extend its utility. The first cab off the rank was giving users access on their mobile devices. This mobile feature-set had several benefits, such as letting workers simply ‘tap on’ when in proximity to their assigned supervisor. The health and safety messages for the day were the first notifications the worker saw. In addition to the standard functionality, the supervisor could approve overtime in real time, allocate a worker from a different home base and permit exit from site to run errands without losing billable time.
“The best results are accomplished when
all parties see the win-win”
Having all the work on the app also meant supervisors and package owners could assign workers to specific work orders and operations on the fly. The technology also allowed for the infield recording of the start and end times of each work order, making it available for later analysis.
We figured that these tens of thousands of entries, when put under the scrutiny of some machine learning and artificial intelligence programs would provide whole new levels of insight into how to better plan and perform the work.
A powerful way of integrating the technology into the overall strategy was to continuously link it back to the organisational goal. At every decision point, we asked ourselves the question: ‘Is what we’re about to do going to get us closer or further from our goal of safely and reliably being in the bottom quartile of our commodity’s cost curve?’ If the answer was in the affirmative, generally speaking, we would do it. But, we also became adept at asking the question in a different, more pro-active way: ‘What action can we imagine that will make our new technology even more powerful in supporting our organisational goal?’ Sometimes the answer was simple, and we could just do it. At other times, the technology unknowns, cost and risk meant we had to go away and think more deeply. All the while, though, we were informed by how the decision could create a positive business case.
In summary, the use of the six questions proved extremely valuable in two significant ways. Firstly, it provided all the players with a framework for thinking about the technology journey in a comprehensive manner. Secondly, we avoided the trap of delivering a less ambitious solution by engaging deeply with each of the questions and recognising how much more value there was to realise when we were bold enough to lift our vision to incorporate the systems view.
Read Part I: Six Questions to Judge Technology
» Learn how our our services can help you.
» Return to view all articles.
» SUBSCRIBE TO NEWSLETTER and never miss an article
[Background image: Alumina Smelter]
All improvement is change, but not all change is an improvement. We can optimise sales, or marketing, or manufacturing, or supply chain, or HR, or IT. Indeed, you can optimise any subsystem of your organisation. But how will it affect your business outcome? (more…)
Discover better ways to do better work.
We alternate our own actionable articles with three relevant links from other authorities.We’ll only use your email address for this newsletter. No sales calls