Open source business intelligence tools deliver the goods for GeoPost

There’s a strong case to be made for business intelligence investments in the transport sector. And for one parcel carrier, the BI business case was based on open source BI tools.

This article can also be found in the Premium Editorial Download: IT in Europe: Data governance key to MDM takeoff

From A to B. It sounds a straightforward proposition. But with a myriad of obstacles in between, logistics and distribution firms make an art out of getting it right for the right price. Journey times, legal requirements, traffic congestion, fuel consumption, maintenance costs and customer service quality all vie to influence the bottom line.

In these circumstances, you might think top-tier business intelligence (BI) vendors would be first in line to help transport companies negotiate the right route. However, best in class is not always the best fit with business needs, as one of the UK’s leading express parcel carriers has found.

GeoPost UK, which is owned by French operator La Poste, was faced with a proliferation of performance data after a system that generated Excel spreadsheets from a multidimensional database was overtaken by business requirements. The company’s use of data had become fragmented, and users wanted to be able to find more detail on particular metrics, such as service problems or revenue blips.

Vernon Adams, GeoPost UK’s head of technology architecture and strategy, said: "We did have a good information system from the 1990s from a business intelligence tool. ... We fed that data warehouse, and users were given access via Excel. What happened then was that the user base built up their own information systems from that summarised database, and of course, as the business has grown they want to go a bit lower and understand what has driven particular figures

“They are all producing information and applying their own filters and exceptions, and of course, you are getting different picture from different parts of the business about the same information. So the business bought into the goal of having a single version of the truth, and that we invest in a new business intelligence system to enable reporting and analysis."

What wasn’t foreseen, though, is that the BI system would end up being based on open source business intelligence tools.

Top-tier BI software gives way to open source BI tools

Initially, several leading BI vendors were invited to present their software to the board, using GeoPost’s own data to create executive dashboards. However, the results were underwhelming, particularly as the most costly came with a £1.5 million price tag.

GeoPost could have afforded the required investment, Adams said, but the board members didn’t see value in the proposals.

“I could see the guys in the board room thinking, ‘This is nothing special here; I’m not able to analyse this, and there’s so many different tools,’” he said. “You start to look at the price, and with all the training, it was mounting up.”

Next port of call was open source BI tools. Adams found that although they didn’t always have the bells and whistles of the top proprietary BI software suites, they were able to perform the same core tasks. And while the big-name suites had been put together from different products, often following acquisitions of vendors, the open source BI offerings were built through open projects that Adams said were easier to deconstruct.

This allowed the GeoPost IT team, working with an open source management consultancy, to build its own bespoke dashboards, based on years of experience with internal business needs, and plug in open source BI and database software under the bonnet.

“I’ve been with the business a long time. I understand all the metrics and the management reports,” Adams said. “It took some organisation to assimilate it, but we created a set of dashboards where 700 people are all looking at the data in the same way.”

As a result, GeoPost UK’s business analysts can see the direct links between operations performance measures, such as delivery times, and business metrics, such as profitability and level of service.

Each customer is quoted a delivery price based on a particular demand profile volume of parcels, together with size and delivery distance, and so on. If the customer’s use of the parcel service deviates from the profile, analysts can see that from the previous day’s data, enabling them to adjust their way of working based on the open source BI findings.

“We’ve introduced a number of operational processes which are driven now from the dashboards,” Adams said. “It’s had a significant improvement in the way that we manage at depot level and helped from a profitability and account retention point of view.”

Key to getting GeoPost UK’s managers to use the open source BI system was engagement right from the top. “The CEO is the main driver behind it. If we didn‘t have the CEO buy-in, it would have been seen as an IT project,” Adams said.

High fuel costs stoke broader interest in BI

The case for investment in BI technologies in the transport sector is compelling, not least because high oil prices have exacerbated one of the greatest cost pressures in the industry – that of fuel.

Gartner research vice president Andreas Bitterer said the proportion of turnover spent on fuel in the logistics sector is considerable. But he added that the fuel bill could be cut dramatically by recording operating data with on-board digital tachographs and analysing the data with BI tools,whether they’re commercial products or open source BI software.

 By assessing the influence of subtle differences in acceleration, breaking and engine revs, the firm was able to nudge driver behaviour by incentivising more efficient operation of vehicles. “I’ve spoken to a trucking company that saved $60 million in fuel,” Bitterer said.

“I’ve spoken to a trucking company that saved $60 million in fuel,” Bitterer said. By assessing the influence of subtle differences in acceleration, breaking and engine revs, the firm was able to nudge driver behaviour by using incentives to encourage more efficient operation of vehicles.

“That kind of information is quite interesting because it has a huge impact on the fuel consumption of the truck,” Bitterer said. “If you are driving in a certain style, tyres and fuel last longer. Times by 10,000 or 50,000 trucks, that’s a lot of money. That’s total BI: using information to make business decisions.”

There’s an even wider scope for BI to help transport firms with more sophisticated decision making. That includes real-time route optimisation, by which data about the locations of vehicles can be combined with traffic flow data to assess if changes to routes are beneficial to meet last-minute orders while trucks are in transit.

“I can decide that it makes sense to do a little detour and add a little time and cost but still be beneficial overall,” Bitterer said. “That’s quite a bit of algorithms that need to kick in to find out how we optimise the route. Not many firms do this, but those that do benefit from data mining and analytic capability.”

Beyond that, BI tools can help transport firms find the best locations for warehouses and distribution hubs by combining transport data with information from sales, human resources and other internal systems as well as external data such as property prices.

With greater computing power in cab, combined with satellite location data, there is potential for greater application of BI technologies to help firms improve customer service, reduce cost and lower carbon emissions, particularly with the larger firms in the sector, Bitterer said.

Read more on Business intelligence and analytics