Inspections Roadmap highlights and priorities
When an organization has multiple questionnaires, they often share identical sections. Components library allows to quickly reuse and manage these sections, which can consist of questions, groups of questions and corresponding logic.
The information collected by field teams can be reviewed and evaluated with our web-based analytics tools. We aim to provide a clear-cut overview of the collected data for a specific questionnaire and also show statistics and insights for the entire set of available data.
We want to make the Inspections setup as easy as possible for you. Whether you already have an existing project you’d like to extend with Inspections or you are starting from scratch, enabling Inspections and the necessary configuration within any mobile project will become a matter of few clicks.
All features and details
As part of the Analyst tool, users will be able to list, filter, view, and read the results of completed questionnaires. This allows users to focus on certain answers and effectively explore possible correlations. Once the results are filtered, the completed questionnaire can be viewed in the web application on the backend, so analysts and managers can track the progress.
Resco Inspections can utilize existing business data by connecting to external data sources. The setup for these sources can be simplified, focusing on different aspects of the setup and demo experience.
A wizard-like experience will make it easier to enable Inspections within any Resco mobile project, existing or new, in just a few steps:
- Enable Inspections in your project
- Add questionnaire views (Answered, Templates, cells)
- Create metadata connection between questionnaires and entity (e.g. appointment, N:N entity)
- Add a questionnaire as an associated tab to entity
- Add a Questionnaires item to Home
- Add a Questionnaire form/viewer to the browser app (Microsoft Dynamics 365/ Salesforce)
Many inspection scenarios can benefit from voice control of the application. If the inspector’s hands are occupied, the application will read out loud the content of a questionnaire form and provide possible options. Then, it will wait for a voice input – for the user to dictate the answer which is then entered into the questionnaire automatically.
The voice control mode will be enabled by a gesture, command or an activation keyword. Additional keywords will be available for e.g. repeating the text, repeating the value of an answer or for skipping to the next item. Finally, it should be possible to enable the text to speech and speech to text options of voice control separately, based on the requested use case.
Components library – Composite components
Snippets streamline the process of reusing portions of a questionnaire, but they lose connection to the original snippet the moment they are pasted. This means that modifying the original and thus making a change to all occurrences of a snippet pasted into multiple questionnaires is not possible.
Composite components resolve this issue. They remain connected to the original component, so all future updates to the component are reflected in all questionnaires, where this component is used. A composite component is a ‘black box’ which has inputs and outputs. Once inserted into a Questionnaire it remains non-modifiable, so future changes to the component are possible only in the Component library.
Localization (Language support)
In an environment where an organization operates in multiple countries, or in a country with multiple official languages, it is usually required to manage the same questionnaire, but in different languages. This means having the same questions and the same logic, but the texts – question labels, descriptions and answer options, etc. should be available in multiple language variations.
Our goal is to incorporate this localization directly into the Designer, so different text versions can be handled there. Also, an export/import function will be available for efficient translation setup.
Pages and decision tree
Many questionnaires are structured as process dialogs or wizards. They take the inspector through the process step by step, displaying one or more questions at each step and each answer possibly affecting which step in the process will follow.
In Resco Inspections, we will simplify the modelling of this approach with Pages. At any given moment, only one page is displayed. Each page can display one or more questions and question groups. Based on a precisely defined decision tree, a set of answers on one page might lead to a certain page being displayed next. If the answers are different, a different page will be displayed next.
The decision tree driving the process is a key element and will be displayed in a specific interface allowing visual representation and editing – so the relationships between pages are immediately apparent.
Additional question types
Different organizations have different requirements. To capture the different types of data these organizations need, we’ll provide additional question types such as:
- Rating to be able to rate a product or a service in a visual way (e.g. 1-5 stars)
- Table to be able to select multiple options from a database table (e.g. select all products the customer is interested in)
- Progress bar within the questionnaire to show how much is still left to be answered
- Editable static image for inspectors to highlight any damages or flaws (e.g. to indicate a damage on a schema picture)
- Augmented reality item (e.g. location-based annotations)
- Map item question for selecting and recording a location on a map
- Numeric range question with custom values in a fully configurable slider
- Relationship question to define a relation between a questionnaire and an entity
The process of designing a questionnaire can result in complex inspection questionnaires, with multiple different paths and complex logic. The ability to immediately test the questionnaire as a preview – as the user would see it in the mobile application – will make the design process significantly more effective. The preview will be part of the browser experience, directly launched from the Designer.
Building on top of voice control, we plan to incorporate conversational interfaces for filling out a questionnaire. Similar to standard voice control, a conversational interface allows the user to provide answers into the questionnaire. The main difference is that instead of using keywords, the user interacts with the questionnaire using standard phrases (such as “I’m opening the machine.”). The conversational engine detects intent in the phrase which corresponds to a question or a step (“Open the machine”) and automatically fills in the answer (“Done”). Additionally, the engine can continue the interaction, outlining answer options or next questions (“Is the machine operational?”), similarly to chatbots.
There are many different scenarios where the inspector’s hands are not free and many of them would benefit from displaying information in the user’s field of vision. Augmented Reality devices, such as Microsoft HoloLens and others, could prove useful. An application UI/UX optimized for this type of devices will be developed. Users will be able to visualize inspection questions one by one (question after question) and theit head movement will indicate an answer (yes/no) or launch voice recognition.
Images represent a valuable data set, especially when combined with advances in the field of artificial intelligence. An existing database of pictures of products, assets, or machines can be used in conjunction with image recognition to identify the object captured in a photo. This can be then used in lookup questions to select the corresponding record (product, asset, etc.) by taking a picture and/or to start the entire process by it. Utilizing Azure Machine Learning (Azure Custom Vision) should provide a convenient technological basis to build this feature.
Wearables offer a quick way to improve the efficiency of inspections. A smart watch can be used to quickly check in or check out of a job. Integration with voice control will allow answering simple types of questions that don’t require complex inputs.
Data Layer improvements
The data-driven architecture of questionnaires allows great flexibility without the need to create metadata on the backend system. However, representing each question and answer as a record might be inefficient for some use cases. Therefore we will introduce the following options:
- Optimized storage – Instead of storing each answer as a record, we will store all answers from a questionnaire in one record as a JSON object. This will speed up the synchronization process and streamline data storage usage.
- Hybrid storage – Similar to Optimized storage, Hybrid storage will store specified answers from a questionnaire as a JSON object. The remaining answers will be stored as records, so they can be easily used with various reporting tools.
Data collected using Resco Inspections can be analyzed using the standard analytics tools available in the connected backend (Microsoft Dynamics 365, Salesforce, Resco Cloud), but for immediate and quick analysis Resco provides the Automatic dashboard. For each Questionnaire template, this dashboard offers an overview for completed questionnaires and collected data. For each question a corresponding representation is selected (a chart, a list, etc.) displaying the ranges and values of answers. Without any additional setup, Automatic dashboard allows direct insight into your inspection process, where filtering on users, dates and additional criteria allows for effective drill down into data.
Resco’s Roadmap might be subject to change. The outlined features may be modified or updated before commercial release. Resco makes no warranties with respect to the information provided.
Last updated: 21.5.2019
The work on this feature has already started. The expected delivery date in next release (usually March, June, September or December), but if the feature is more complex it might require two release cycles to be completed.
The feature has been implemented and will be released in Resco’s forthcoming release (next: June 2019).
This feature is on Resco’s roadmap and is waiting to be picked up by the development pipeline to get implemented.
The feature has been released in last Resco’s release (Version 12.0 – March 2019).