To optimize your field operations, you need good planning software but – first and foremost – you need data. Which data? Where to find it? How to ensure its quality? And how to go to the next level by mobilizing additional tools and data?
Implementing a planning and field service optimization software product is a strategic project from which you rightly expect operational efficiency enhancements and greater customer service capability. You can select the best software on the market, but you find it very hard going to achieve these objectives if you fail to tackle the question of what is fueling your software head-on: data, namely all the information it cannot do without if it is to create realistic schedules and routes.
Beware: this is not about incorporating vast quantities of unsorted data on the basis that it is better to have too much than not enough. The challenge is to incorporate the data the software needs to operate and achieve your goals.
Think “data” right from project inception
The more processes are computerized, the more you think that data is available and, moreover, of good quality. This is far from always being the case. Because of this belief, data is often addressed in a second stage, after the functional part of the project. And it is just when you think that your goal is within touching distance that you discover that the data is inaccessible, incomplete, obsolete, or even non-existent!
If you want your software to deliver all the added value it can provide, if you want to put your new planning tool into production within the anticipated timescales, you must conduct two activities in parallel:
- a functional activity, consisting in defining what users need, and configuring the solution so as to meet these needs, without falling into the trap of ultra-customization;
- a data focused activity which aims 1/ to verify that all the data required by the functional choices exists; 2/ to audit the quality of this data; 3/ to put in place an action plan to improve and maintain data quality – all this within a time-scale compatible with that of the functional part of the project.
If you become bogged down with this second activity, or if you underestimate its scale, not only will you delay the deployment of your software, but you risk delivering a tool that will disappoint planners and field teams alike. The former because the solution will not deliver on its promises of simplification and automation; the latter because the schedules and routes generated by the software will not adequately take account of the constraints inherent in their business and real-life field circumstances.
The 3 indispensable data categories
In all Field Service Management solution deployment projects, three categories of data determine the quality of operational planning and the accuracy of the routes calculated by the software engine:
1- customer /callout site location data. Without accurate location data it is quite simply impossible to create optimized routes. Your address files (originating from your CRM, ERP, or other database) are the basis on which the software or included in the routes. Do you think the addresses in your customer files are flawless? The only way to be sure is to run a geocoding test, which will identify the addresses that cannot be associated with exact geographical coordinates because they are you are using (neighborhoods under construction, newly created roads…). The higher the exact address geocoding quotient, the more accurate and relevant future routes will be. It is therefore very much in your interest to run this test and plan a cleaning up/location data quality upgrading operation before importing them into the system.
>> Ensure that the addresses you are incorporating are indeed those of the locations where your teams will be calling and not the billing addresses…
>> Also include the addresses of your agencies, shops, depots and all those locations likely to be a departure or arrival point in your future routes.
2 – Service call duration-related data – knowing whether it will take a quarter of an hour or an hour to provide a service is obviously a crucial item of information in creating sustainable schedules and balanced routes. Because service provision generations durations can vary considerably depending on the service in question, you need to provide the software with a detailed typology associating an average duration with each type of call. If no such reference table exists, you will need to create it, typically from information that (normally) features in service call reports (service provided, service call start and end times). If this information is unavailable or if there is insufficient historical data to be statistically representative, you are left with only one solution: ask your technicians, calculate the average durations from their answers and compile all that in an Excel spreadsheet. Your software will then be able to model the durations and take account of them when calculating routes.
>> Creating this reference table is an indispensable prerequisite to using the software. Its accuracy has a major impact throughout the organization. If you model services at 30 minutes which actually require 20 minutes, this very tangibly means fewer calls in a day, fewer customers invoiced, longer appointment times for customers, less productive days for your field operatives.
>> Do not confuse call duration and access time to the equipment/appliance the technician is required to service. For example, the time it takes to access a private individual’s boiler will be very short once the technician has arrived on site. On the other hand, if he is calling on a hospital group or a sensitive site with numerous access checks, he may need 10 or 15 minutes to reach the equipment to be maintained. Certain activities therefore require the two items of information to be modelled – the time to access the equipment and the effective call duration – to calculate accurate and reliable times.
3/ Data on operatives’ skills – In the equipment installation, repair, and maintenance trades, and in the survey trades, calls must at all costs be assigned to individuals possessing all the required qualifications and authorizations. You can assume that planners are well acquainted with “who can do what” and that their experience will enable them to allocate assignments to the appropriate operatives. The thing is that they do this manually, on an ad hoc basis, whereas your software can automate this part of their job and save them a lot of time. It is therefore highly recommended, from the outset of the project, to create a skills reference table, assisted by the Human Resources Department, planners, and the operatives themselves. This reference table will enable anyone able to perform a given service to access the system and be allocated only those assignments that match his or her skills or area of expertise.
>> It is also recommended that a typical schedule be developed for each operative for a one- or two-week period to take account of individual circumstances (part-time, negotiated hours). This may require considerable effort but typically HR departments have all this information. If you take the trouble to incorporate this data, the software will know that it cannot allocate an assignment to such and such a technician on Wednesday, but that another technician can carry out calls until 19:00 on Fridays, unlike his colleagues with the same skills, but who finish at 17:00.
Improving accuracy and efficiency over time
We saw the 3 priority data categories you need to prioritize to successfully deploy your FSM solution. The data issue does not stop there: your activities will develop, you will offer new services, there will be changes in the ranges of appliances you install or maintain, colleagues will relocate or leave the company, others will arrive, you will win contracts and new customers….
To ensure your software is constantly adding value, you need to provide it with this information and enable it to work with up-to-date data at all times in the three fundamental categories. Although certain amendments/updates can be done manually, it is far more efficient to create bridges between your business solution and the main systems from which this data is drawn:
- the CRM and/or ERP which (theoretically) holds the customer reference data, in particular their addresses;
- the HR system, which centralizes co-workers’ statutory and contractual information, together with training and skills data. and where this information is updated;
- The maintenance management system (GMAO), which centralizes data on each component of your customers’ installed fleet, contracts entered into, and the service level commitments associated with these contracts (statutory number of calls, customer service response times, etc.)
and these systems automatically update your reference tables, as they occur or periodically. This data exchange method is also how your model can be enhanced by incorporating additional information that will be taken into account when calculating routes and which will improve its accuracy – for example, dates when co-workers are on leave or else the opening hours of stores you are delivering to, or the air conditioning systems of which you are maintaining. This will avoid the engine scheduling visits on days when the stores are closed or assigning calls to co-workers who are absent.
Not all this information needs to be held in existing databases, but you can incorporate it by asking your operatives to upload it using their mobile business application. This is the same application you can use to fine tune, and if necessary correct, call durations as operators systematically use it to record each assignment’s start and finish time.
Everyone benefits if you regularly enhance data, if you put in place processes ensuring it is accurate, in the correct format, and up to date:
- your planners, because they make fewer manual changes and corrections, and are better able to anticipate events;
- your technicians, salespeople or delivery personnel because they know that their route plans take account of their individual constraints, geographical constraints, and that they are based on realistic travel and call duration times;
- your customers, because you can give them accurate times, and deliver on your commitments.
Obviously, you can take your FSM solution’s modelling and the service your field teams receive even further. For example, you can link each assignment with the list of spare parts and equipment to take to be sure of being able to complete the assignment in a single call. You can harness machine learning to create this reference table from your call history and stock management system data. Provided you give it enough usable data that is representative of the calls that your teams perform at customers’ premises, the algorithm will extract from your historical data the list of parts needed for each type of call and refine it based on the parts actually used during subsequent operations.
Do you dream of having a solution offering you these services? Start with the basic building blocks, making sure you can provide your planning software with the 3 indispensable data categories. And if you have any doubts about the quality of your location data, put them to rest by giving our consultants the job of auditing your databases. You will save time on the subsequent stages of your project and be able to progress to achieving value-added services faster than you imagine.